Rent migrating upstream
countervailing a named mechanism, not a conclusion
These names pay the fattest supplier margins in the model: what their suppliers earn is their cost line, so rent is being taken out of them from above. A bottleneck one layer up is indistinguishable from a cost problem on the income statement.
The path
| node | effect |
|---|---|
| Google TPU (Alphabet) | -0.50 |
| Together AI | -0.50 |
| NVIDIA B200 | -0.50 |
| NVIDIA H200 (SXM) | -0.50 |
| Nvlink | +0.36 |
| Fluidstack Ltd. | -0.50 |
| Mistral AI | -0.50 |
| NVIDIA H100 (SXM) | -0.50 |
| NVIDIA GB200 (per-GPU in NVL72) | -0.50 |
| NVIDIA Vera Rubin VR200 (R200) | -0.50 |
| LPDDR (low-power DDR) | -0.50 |
| HBM5 | -0.50 |
coverage
Computed from input_cost_pressure, a weighted blend over supplier margins. 903 of 1,848 supply edges carry no disclosed weight, so this blend is partly over a 0.5 default — read the ORDERING, not the level.
Arguments about this mechanism
None yet. No thesis names this mechanism’s subject, which is a gap in the bank rather than a fact about the mechanism.
Arguments that run through it
These name a node on the path rather than the subject, so they pass through this mechanism without being about it. Shown separately and never graded as claims about it — hubs above the graph’s own 90th-percentile degree are excluded, or every argument touching NVIDIA would attach to every mechanism NVIDIA touches.