A paywall subscription to the Boston Globe will run me (long term) $7/week. That's $0.30/week less than they would charge me for a daily physical paper.
But in order to make that a strong argument, you are implicitly assuming that the marginal cost of the physical version is significant. It's probably easy to argue that it is significant compared to the marginal cost of online distribution, but is it significant compared to the capital costs?
In the same vein, what are you including in your model of marginal cost? Are you including printing press and datacenter operations in your mental calculation, or only the cost of distribution itself? Because with those, the marginal costs are much closer to eachother than without -- although they're possibly still an order of magnitude apart, I really have no idea.
But in order to make that a strong argument, you are implicitly assuming that the marginal cost of the physical version is significant. It's probably easy to argue that it is significant compared to the marginal cost of online distribution, but is it significant compared to the capital costs?
In the same vein, what are you including in your model of marginal cost? Are you including printing press and datacenter operations in your mental calculation, or only the cost of distribution itself? Because with those, the marginal costs are much closer to eachother than without -- although they're possibly still an order of magnitude apart, I really have no idea.