The most exciting aspect of MIT Press's [Direct to Open](https://direct.mit.edu/books/pages/direct-to-open-fees) OA book initiative is that library-participant fees are set, in part, according to ability to pay:
> ## To determine fees that allow the offer to reach the revenue target, the Press:
> * Assessed the total universe of academic institutions—by type, size, and region—that could participate in a collective open offer;
> * Estimated (based on past demand) the percentage of institutions in each segment expected to commit to one or both subject collections;
> * Estimated the relative value of each collection based on number of titles;
> * **Indexed the fee tiers based on the median acquisitions budget for each tier**; and
> * Set provisional fee levels sufficient to achieve the revenue target.
The [MIT model](https://direct.mit.edu/books/pages/direct-to-open-fees) was designed in collaboration with Raym Crow—arguably the leading figure in the non-APC OA funding world, going back to a [classic 2006 paper on publishing cooperatives](https://firstmonday.org/ojs/index.php/fm/article/view/1396). He helped *Annual Reviews* with its [subscribe-to-open model](https://onlinelibrary.wiley.com/doi/full/10.1002/leap.1262), and, since then, has been [working with MIT Press](https://mitpress.mit.edu/blog/starting-scratch-finding-place-open-access-monographs) on this Arcadia-funded project.
The best feature of the [MIT model](https://direct.mit.edu/books/pages/direct-to-open-fees) is its reliance on ability-to-pay (as proxied by acquisition budget) in setting membership fees. Not institution size, not authorship counts, nor other wealth-agnostic factors.