Geopolitics, Geoeconomics, and Sovereign Risk: Different Shocks, Different Channels

This paper distinguishes between geopolitical and geoeconomic shocks by demonstrating that while geopolitical risks directly reprice sovereign default risk, geoeconomic shocks transmit through monetary policy and the global financial cycle, creating a "scissors pattern" in sovereign CDS spreads that necessitates different policy responses for liquidity provision versus persistent risk premia.

The Gist

Imagine the global economy as a massive, interconnected neighborhood where every country is a house. The "price" of borrowing money for these houses is called the Sovereign Spread (or CDS spread). If the price goes up, it means the neighborhood thinks that house is risky and might not pay back its loans.

This paper asks a simple but crucial question: When bad news hits, why does the price of risk go up? Is it because the house itself is damaged, or because the whole neighborhood is panicking?

The authors, Alvaro Ortiz, Tomasa Rodrigo, and Pablo Saborido, discovered that there are two very different types of bad news, and they break the neighborhood in two completely different ways. They call this the "Scissors Pattern."

Here is the breakdown in simple terms:

1. The Two Types of Bad News

Think of bad news as coming in two flavors:

• Flavor A: Geopolitical Shocks (The "War" News)
  • Examples: A war starts (Russia-Ukraine), or a terrorist attack happens.
  • The Effect: This is like a fire starting in a specific house. The risk is direct. The house itself is in danger. The market immediately says, "That specific country is risky because of the fighting."
• Flavor B: Geoeconomic Shocks (The "Policy" News)
  • Examples: A country announces new tariffs, a confusing election, or a trade war.
  • The Effect: This is like the neighborhood association changing the rules or the bank changing interest rates. The house isn't on fire, but the environment around it has become scary. The risk comes from uncertainty and fear, not from the house actually burning down.

2. The "Scissors" Pattern (The Core Discovery)

The authors found that these two types of news move the "risk price" in opposite directions using different levers. Imagine a pair of scissors opening and closing:

• When War Happens (Geopolitical):

  • The Direct Lever (The Blade): The risk price for the specific country shoots UP immediately. The house is damaged.
  • The Global Lever (The Handle): Surprisingly, the global financial system sometimes acts as a shock absorber. Because the world is so connected, global investors might temporarily calm down or shift money in a way that pushes the risk price DOWN slightly for everyone else.
  • Result: One part goes up, the other goes down. They "scissor" against each other. The net result is a messy, complex risk increase.

• When Policy Changes Happen (Geoeconomic):

  • The Direct Lever: The specific country's house isn't really damaged yet, so this lever barely moves.
  • The Global Lever: The whole neighborhood gets nervous about the new rules. The "Global Financial Cycle" (the mood of the whole street) goes wild.
  • Result: The risk price goes up because of global panic and uncertainty, not because the specific house is broken.

3. The "Gravity" Rule vs. The "Global" Wave

The paper also looked at who gets hurt and how far the damage travels.

• War (Geopolitical) follows "Gravity":
  • Imagine dropping a stone in a pond. The ripples are strongest right next to the stone and get weaker as you move away.
  • If a war starts in Ukraine, countries nearby (Germany, Poland) feel the risk immediately. Countries far away (Australia, Brazil) feel much less of it. The damage fades with distance.
• Policy Changes (Geoeconomic) are like a "Global Wave":
  • If the US announces a confusing new tax or election result, the fear hits everyone at the same time, regardless of distance. It's a wave that washes over the whole neighborhood simultaneously.

4. Why This Matters for Policymakers (The "Fire Extinguisher" Problem)

This distinction is huge for Central Banks and governments.

• If the problem is the Global Lever (Geoeconomic):
  • The Central Bank can act like a firefighter with a hose. They can provide liquidity (cash) to calm the neighborhood down. Since the house isn't actually burning, just the fear is high, giving people cash usually fixes the problem.
• If the problem is the Direct Lever (Geopolitical):
  • The house is actually on fire. Giving the house a bucket of water (liquidity) won't stop the flames. The risk is real because of the war or conflict.
  • The Solution: You don't need a fire hose; you need diplomacy to stop the fighting, or fiscal fixes to repair the house. Money alone cannot fix a war.

5. How They Found This (The "Magic Decoder")

The authors didn't just guess. They used a super-smart computer (Machine Learning) to read thousands of news articles every day in local languages.

• They built a "decoder" that could look at a country's risk price and say: "Okay, 40% of this risk is because of the war (Direct), 30% is because the global mood is bad (Global), and 30% is because people are confused about taxes (Uncertainty)."
• They tested this decoder against four major recent events (Russia-Ukraine, Hamas-Israel, the US Election, and US Tariffs) and found their "Scissors" theory held up perfectly.

The Bottom Line

The world is getting more dangerous, but not all dangers are the same.

• War breaks the house directly, and the damage is felt most by the neighbors.
• Policy confusion scares the whole neighborhood, but the houses are still standing.

To fix the problem, you have to know which one you are dealing with. You can't fix a war with a cash injection, and you can't fix a panic with a treaty. You need the right tool for the right job.

Technical Summary

1. Problem Statement

The paper addresses a critical gap in the literature regarding sovereign risk transmission: while it is established that both geopolitical and geoeconomic shocks widen sovereign credit default swap (CDS) spreads, it remains unclear whether these two shock types operate through the same transmission mechanisms. Existing literature often treats "uncertainty" as a monolithic driver. The authors argue that geopolitical shocks (e.g., military conflict) and geoeconomic shocks (e.g., trade wars, policy uncertainty) fundamentally differ in how they reprice sovereign risk, necessitating distinct policy responses. The core question is whether these shocks transmit through a direct repricing of default probability or indirectly through global financial conditions and uncertainty channels.

2. Methodology

The authors employ a three-layered empirical framework combining machine learning, game-theoretic decomposition, and causal inference:

A. Data Construction

• Sample: A daily panel of 42 advanced and emerging economies from January 2018 to July 2025.
• Dependent Variable: 5-year Sovereign CDS spreads (standardized).
• Predictors:
  • Global Financial: VIX and U.S. 2-year Treasury yield.
  • News-Based Indicators: Constructed from the GDELT database using local-language media (not global outlets) to capture domestic risk perceptions accurately. These include Geopolitical Risk (GPR), Economic Policy Uncertainty (EPU), Trade Policy Uncertainty (TPU), and sentiment indicators for economics, interest rates, and political tensions.
• Preprocessing: Variables are smoothed with a 28-day moving average and standardized.

B. Machine Learning & Shapley-Taylor Decomposition

1. Forecasting Model: The authors compare 15 model classes (linear, penalized, tree-based, neural networks). They find that nonlinear models (specifically a Multilayer Random Forest) significantly outperform linear models, with news variables improving out-of-sample forecasts by 15–19% in nonlinear settings versus 5–9% in linear ones. This suggests transmission operates through interactions and threshold effects.
2. Channel Decomposition: Using Shapley-Taylor interaction values, the authors decompose the predicted CDS spread into four distinct channels for every observation:
   • Direct Channel ($\phi_{dir}$): The driver's own contribution (e.g., default risk repricing from conflict).
   • Global Financial Cycle (GFC) Channel ($\phi_{GFC}$): Interactions with VIX and US2Y (risk appetite and monetary conditions).
   • Uncertainty Channel ($\phi_{UNC}$): Cross-reinforcement among policy uncertainty measures.
   • Local Channel ($\phi_{LOC}$): Interactions with domestic sentiment, political tensions, and regional structure.

C. Econometric Validation

To ensure the channels are not artifacts of the ML model, the authors use Panel Local Projections (Jordà, 2005) with two identification strategies:

1. Full-Sample Innovation: Using AR(5) residuals as shocks (finds muted effects due to averaging).
2. Narrative Identification: Using event dummies around four specific crises:
   • Russia–Ukraine invasion (Feb 2022).
   • Hamas–Israel conflict (Oct 2023).
   • U.S. Presidential Election (Nov 2024).
   • "Liberation Day" Tariff announcement (Apr 2025).
3. Placebo Falsification: Comparing event responses against 100 random non-event dates to verify statistical significance.

3. Key Contributions

1. The "Scissors Pattern": The paper identifies a distinct transmission mechanism where geopolitical shocks cause the Direct channel to rise (repricing default risk) while the GFC channel moves in the opposite direction (providing a transient offset). This creates a "scissors" effect where the total spread widening is smaller than the direct impact alone.
2. Taxonomy of Shocks:
   • Geopolitical Shocks: Activate the Direct channel (default repricing) which decays with geographic distance (gravity model).
   • Geoeconomic Shocks: Bypass the Direct channel, transmitting primarily through the GFC and Uncertainty channels.
3. Methodological Innovation: The integration of Shapley-Taylor decomposition with narrative local projections allows for the separation of nonlinear transmission channels and their subsequent causal validation, bridging the gap between ML prediction and structural econometrics.

4. Key Results

A. The Scissors Pattern (Geopolitical vs. Geoeconomic)

• Russia–Ukraine (Geopolitical): The Direct channel jumped to +0.10, while the GFC channel dropped to -0.12. The negative GFC response (global financial system absorbing shock) partially offset the direct risk repricing.
• U.S. Tariffs/Election (Geoeconomic): The Direct channel remained near zero. Instead, the GFC channel declined (due to expected monetary easing) and the Uncertainty channel rose.
• Persistence: The Direct channel (fundamental repricing) is persistent (retains ~2/3 of impact after 3 months), whereas the GFC channel is mean-reverting and fades faster.

B. Cross-Sectional Fingerprints

• Distance Decay: Geopolitical Direct effects decay significantly with distance from the conflict zone (e.g., $R^2=0.35$ for Russia–Ukraine). Countries closer to the epicenter face higher direct repricing.
• Global Synchronization: Policy-uncertainty shocks (like the U.S. Election) activate the Uncertainty channel broadly and simultaneously across all countries, regardless of distance.
• Originator Penalty: In geoeconomic shocks, the country initiating the shock (e.g., the U.S. in the tariff episode) sees spread widening through the Local channel (domestic uncertainty), even as the GFC channel compresses spreads for the rest of the world.

C. Econometric Validation

• Narrative LPs: Recovered the "scissors" pattern at the 1% significance level for the Russia–Ukraine event.
• Prediction Accuracy: 15 out of 16 event-channel predictions were confirmed (e.g., Geopolitical events activate Direct; Geoeconomic events activate Uncertainty/GFC).
• Placebo Test: All four episodes produced channel divergences exceeding at least 83% of random non-event dates, with the specific channels exiting the envelope differing by shock type as predicted.

5. Significance and Policy Implications

• Policy Design: The distinction has profound implications for central banks.
  • If spread widening is driven by the GFC channel (geoeconomic/financial stress), liquidity provision and swap lines can effectively stabilize markets.
  • If spread widening is driven by the Direct channel (geopolitical default risk), liquidity tools are insufficient. The appropriate response requires diplomatic de-escalation, fiscal correction, or institutional strengthening.
• Real-Time Monitoring: High-frequency news-based indicators are crucial not just for signaling risk, but for identifying which channel is driving the risk. As geoeconomic tensions rise, the transmission mechanism shifts toward channels that traditional central bank tools cannot address.
• Theoretical Contribution: The paper provides empirical evidence that sovereign risk is not a monolithic concept but a composite of distinct, state-dependent transmission mechanisms that require different modeling approaches (nonlinear vs. linear) and policy responses.

In summary, the paper demonstrates that geopolitical and geoeconomic shocks are fundamentally different beasts: one reprices the sovereign's survival probability directly, while the other alters the global financial environment and uncertainty landscape. Recognizing this "scissors pattern" is essential for accurate risk assessment and effective crisis management.