Only in a simple market vacuum.
For-profit sites don't accept every single person who applies, even those with 50% acceptance rates. So, yes, there is a large pool from which to draw such persons.
Anecdotal evidence (TD perceptions and recommendations of the supply and demand conference, forum posts on this board, etc.) indicates very few.
Agreed; those are short-term, myopic solutions that would quickly be absorbed by Buck's School of Professional Psychology.
Agreed, and this is in the literature (Stedman et al., Parent and Williamson). But, sites can't really generate internship spots like that. Modifications to the current accreditation practices would be useful, but not at the cost of weakening quality. Pairing the poor match rate site with a local currently unaccredited internship site to help the site get accredited (e.g., grad students paid as RAs to run the required program evaluations for accred. review) is a proposed and potentially viable solution. But, this is a cost to the site, and again there's little intrinsic incentive for a for-profit site to engage in this, as their outcomes meet their goals under the current system. (edit: especially as the developed site would presumably be open to all applicants; generation of captive internship sites is ridiculous. What if UCLA decided one day, "We're only going to take UCLA doc students"? Captive internship site development is also myopic.)