> AAS is a nonprofit society, though, so while there may be some cross-subsidy of other activities those page charges must reflect to some degree the actual costs to maintain the journal.

I've published with ApJ. In late 2016, ApJ charged us around $1500 for a 16-page article, which apart from some minor copyediting, was published as it is. Now, the $1500 fee may not be much for a well-funded research group in a first-world country. However, it is a considerable amount of money for other research groups. If arXiv, which receives over 10,000 submissions and 35 million downloads a month [1] can operate on a yearly budget of $2.5 million [2], I don't see why these journals can operate at a comparable budget.

> Journals without page charges simply pass those costs on to readers (or their libraries), which slows scientific progress.

There is good evidence that scientific progress has slowed down already [3,4] and it has almost nothing to do with lack of open access journals. Science progressed at a much faster rate in early and mid 20th century, where such concepts didn't even make sense. 21st-century science will fair better if funding institutions started caring about the quality of research done.

[1]: https://arxiv.org/stats/main

[2]: https://arxiv.org/about/reports-financials

[3]: https://www.theatlantic.com/science/archive/2018/11/diminish...

[4]: https://www.pnas.org/content/113/34/9384

I would be more sympathetic to cost arguments if they were transparent about it.

In the end I think this starts to raise important questions about sources of costs and benefits of a journal article and who should pay (not about AAS in particular necessarily but in general).