A comparative case study of Nigeria's tariff policy across refined petroleum, automobiles, and rice, 2023-2025 — read against its AfCFTA commitments.
Nigeria doesn't have one tariff policy; it has three, and they point in different directions. This project compares three sectors under materially different tariff treatment over the same policy window and asks a narrow, testable question: does the level of tariff protection a sector receives predict whether that sector actually substitutes imports with domestic production?
The answer is no. The least-protected sector (petroleum, ~0.5% effective tariff) shows the fastest, largest import substitution — driven by a single private refining asset, not policy. The most-protected sector (rice, 70% combined duty) shows the weakest enforcement outcome and remains Nigeria's single most-smuggled commodity. Automobiles, tariffed specifically to reward local assembly (35-40% on imports vs. 0-10% for local-assembly kits), sit at a verified 29.0% aggregate capacity utilization across all 34 licensed assemblers. The variable that actually predicts outcomes isn't the tariff line — it's whether a credible, scaled domestic supply response exists behind it.
Live dashboard: frontend-iota-woad-42.vercel.app
| Format | File | Audience |
|---|---|---|
| Academic paper | paper/nigeria-tariff-academic-paper.md (PDF) |
Literature review, formal methodology, APA references |
| Policy paper | paper/nigeria-tariff-policy-paper.md |
Consulting-style case study |
| Investment memo | memo/nigeria-tariff-investment-memo.md (PDF) |
Capital-allocation read per sector |
| Slide deck | deck/nigeria-tariff-policy-deck.pptx |
13-slide condensed narrative |
| One-pager | paper/nigeria-tariff-one-pager.pdf |
Single-page leave-behind: framework, 3 findings, implication |
| Interactive dashboard | Live site / frontend/ source |
Live charts, built with React + Recharts |
| Video walkthrough script | docs/loom-walkthrough-script.md |
~3-minute recorded overview |
data/raw/ sourced research notes, one folder per sector (+ AfCFTA), every figure cited
data/processed/ cleaned CSVs derived from the raw notes
models/ Python scripts that turn processed data into charts/tables (matplotlib)
outputs/ generated charts (outputs/charts/) and summary tables (outputs/tables/)
paper/ the academic and policy papers, plus the PDF build script
memo/ investment memo
deck/ PowerPoint deck + the pptxgenjs script that generates it
frontend/ React/Vite/Tailwind/Recharts dashboard, data sourced from data/processed/
Every figure traces to a named, dated source, catalogued in data/raw/{petroleum,automobiles,rice,afcfta}/sources.md. Where sources conflicted, both figures are reported rather than silently reconciled — see the "gaps" section at the end of each sector's notes.md. One correction worth flagging: the automotive capacity-utilization aggregate was originally miscalculated at ~24% during initial research; recomputing directly from the primary NADDC table gives the verified 29.0% used throughout.
Regenerate the charts/tables (requires the Python venv with pandas/matplotlib):
python models/petroleum_analysis.py
python models/automobiles_analysis.py
python models/rice_analysis.py
python models/sector_comparison.py
python models/export_frontend_data.py # syncs data/processed/ -> frontend/src/data/
Run the dashboard:
cd frontend
npm install
npm run dev
Rebuild the academic paper PDF:
python paper/build_pdf.py
MIT — see LICENSE.