Third, there are certain categories of employers that are not subject to the cap at all. If you are offered a job at an institution of higher education or a related/affiliated nonprofit organization, a nonprofit research organization, or a governmental research organization, then your petition is not subject to the cap. It is important to note that not all nonprofit organizations are exempt from the cap. Only nonprofits that are affiliated with or related to an institute of higher education, or nonprofit research organizations, are cap-exempt.
The above explains why the number of uncapped H-1B visas is so low. If you aren't looking for a uniquely qualified researcher, then you (the employer) are restricted by the cap. There isn't a useful way to get around the regulation, highlighting the value of Richard Herman's proposal.
The other issue I raised is the lack of mobility for H-1B workers, which may result in exploitation. The lawyer I cited above suggests that the immigrant does have options available to him or her:
There are exceptions to the H-1B cap, but they are very limited. First, if you already have an H-1B visa and are applying to extend your stay or change employers, you are not subject to the cap. Once you have an approved H-1B petition, you do not have to worry about whether H-1B visas are available. Current H-1B holders can apply at any time to extend their stay with their current employer or to change to another employer.
Conceivably, H-1B workers tied to Rust Belt employers could leave the region as long as they could find a sponsor in the desired location. I'm interested in finding data on the secondary migrations of H-1B immigrants. I suspect that the lack of good information is a barrier to geographic mobility and that the potential for abuse (e.g. threatening deportation) still exists. I'm also in the market for arguments against Richard Herman's policy innovation. I've yet to find any substantive objections.