Russia-China Economic Dependency Analysis

Core finding: HHI=0.40 (extreme concentration). RUS 67.5% exports → CHN, asymmetry 7.2x. Event Study post-2022: +53.6% trade (p<0.001). Russia migrated from European to Asian trade bloc (2017→2023).

Have Western sanctions caused irreversible structural dependency of Russia on China?

🌐 Dashboard · Executive Summary · Notebooks

---

Visual Evidence

Figure	Caption
china_share_timeline	China: 5.3% (1995) → 39.4% (2023). Europe: 59.4% → 15.6%.
hhi_timeline	HHI: 0.24 (2007) → 0.40 (2021). Threshold 0.25.
did_treatment_effect	Event Study [+3,+6]: +81.5%. Mean post: +53.6%. R²=0.90.
network_communities_2017_2023	RUS 2017: European bloc. RUS 2023: Asian bloc (incl. CHN).
mdi_timeline	MDI: 2.67%–7.98% (mean 3.89%). Carnegie ~300 M$/mo dual-use.
crink_network_graph	CHN→RUS 245 Bn. Total CRINK: 300 Bn.

---

Analysis

Notebook	Method	N	Result	Test
01	HHI, Gravity OLS	15 yr	HHI=0.404, R²=0.937	Chow p=0.007
02	VAR(2), Granger	96 mo	Brent→Trade p=0.0000	FEVD h=12: 17.2%
03	Event Study	96 mo	+53.6% post-2022	[+3,+6]: +81.5%, p<0.001
04	Louvain	50 countries	RUS: Euro→Asian bloc	CHN share 39.4%
05	MDI	32 mo	2.67–7.98%	Carnegie categories
06	Asymmetry	7 dyads	CHN→RUS 245 Bn	IRN-RUS 4.38

---

Data Sources

Source	Period	Description
UN Comtrade	2017-2024	Monthly bilateral
Harvard Atlas HS92	1988-2024	Bilateral, 200+ countries
FRED	1992-2025	Brent, USD/RUB, USD/CNY
BACI (CEPII)	2017-2023	Reconciled HS6
WITS	2007-2021	HS6 exports
SIPRI	1990-2024	Arms
Carnegie	2024	Dual-use estimates

Processed: russia_china_dependency/02_processed_data/.

---

Methodology

• VAR(2) en dlog. Ordering: Brent → USD/RUB → USD/CNY → Exp RUS→CHN.
• Event Study: phase dummies, ref baseline <−6 mo. OLS, Brent+USD/RUB controls.
• Louvain: edge weight = trade value. Modularity by year.

---

Limitations

• N=96 monthly obs. VAR: 80% power at d≈0.5. Below N=150 for stable IRF (Lütkepohl 2005).
• Post-2022 RUS data: mirror gap CHN-RUS up to −15% (2020).
• DiD SUTVA: sanctions spill to control (IND, TUR).
• Louvain: modularity varies with edge weights.

---

Setup

pip install -r requirements.txt

Run notebooks in order from notebooks/:

1. 01_structural_analysis.ipynb
2. 02_time_series_econometrics.ipynb
3. 03_causal_inference.ipynb
4. 04_network_analysis.ipynb
5. 05_war_dependency.ipynb
6. 06_crink_network.ipynb

License

• Code: MIT
• Reports/figures: CC BY 4.0
• Data sources: original licenses (see russia_china_dependency/01_raw_data/ if SOURCES.md present)