**Claim (regional facet):** Over the next 20 years, US higher-ed disruption is regionally asymmetric, but the asymmetry is demographic-financial, not pedagogical.

1. **Demographic floor:** Post-2010 birth decline shrinks traditional college-age cohorts ~10–15% in the Northeast/Midwest by the mid-2030s, while Sun Belt/SEC-country states gain from births *and* domestic in-migration (zero-income-tax migration, employer relocation). Enrollment follows people.
2. **AI flattens teaching, not place:** As AI makes instruction near-free at the margin, teaching capacity stops being a differentiator. Durable regional advantages become network effects — alumni/employer density in booming metros, cheap land for campus expansion, athletic-media brand flywheel (SEC), and state willingness to fund growth (Texas, Florida, Tennessee vs. Illinois, Pennsylvania budget stagnation).
3. **Predicted 2045 map:** SEC-flagship and Texas/Florida systems *grow* (capacity expansion, satellites, elite athletic-brand gravity); endowed Northeast privates (Ivies, top SLACs) hold as premium scarce brands; the collapse zone is tuition-dependent small colleges in demographically shrinking states — rural New England, Upstate NY, Upper Midwest — where an estimated 15–20% close or merge by 2045.
4. **Interaction with program mix (prior facet):** Region sets the demand floor; program mix sets the slope. A nursing/engineering-anchored regional campus in Ohio can outperform a generic-BA campus in Georgia. "SEC growth vs. Rust Belt decline" is true at the system level but program portfolio can override it at the institutional level.