The problem is that
no news source is good enough to be anyone's
only news source, and since most consumers don't have vendor relationship management software to keep them all straight, we can only keep track of a limited number of subscriptions, across all forms of media, regardless of what each costs in money.
Individually subscribing to 10 different paywalled news sites is not going to happen among the general public. Paying exactly the same aggregate monthly price for all ten, via just one redistributor, is more tolerable, especially if the customer can individually select subscriptions without forced bundling. Subscribe by checking this box; unsubscribe by unchecking it.
If I can sign up online, but have to cancel with at least one phone call guaranteed to last 30 minutes or more, I don't care if it's only $0.05 a month; I'm not buying it. It's not always the money cost. Sometimes it's the cognitive overhead or transaction friction. My use of the Internet lately seems to be governed more by the number of accounts I am willing to manage than by how much I am willing to pay, and the fragmentation is going to kill some of the sites that can't solve that problem.
When the NYT is in full control of its own paywall, you get nonsense like described here: https://dannysullivan.com/new-york-times-subscription-3480 [2014], where it's cheaper to buy the print version and just throw it away to get past the paywall, than to get the unlimited digital subscription. And this is likely fueled by a dark pattern: in order to get a price for print delivery, you have to call a phone number during NYC business hours, whereas you can just click to get an offer for all-digital access. It's only $8/4 weeks right now, which seems like a good money value, but I'm just not in the mood to worry about yet another website login.