Debt, Growth, and the Carbon Lock-In

1. Problem Statement

The paper addresses a critical paradox in climate economics: despite decades of climate policy and significant improvements in energy efficiency (decreasing carbon intensity), global CO2 emissions continue to rise. Standard Integrated Assessment Models (IAMs) typically treat GDP growth as exogenous and fail to model the financial system, thereby missing the feedback loops between credit dynamics and emissions.

The authors argue that financial leverage is a structural driver offsetting efficiency gains. They posit that debt-financed growth creates a "carbon lock-in" mechanism where:

1. Credit expansion stimulates economic growth, increasing energy demand.
2. Growth remains energy-dependent, generating cumulative emissions.
3. The necessity to service debt requires sustained growth, creating a feedback loop that prevents economies from decoupling from carbon, even as carbon intensity declines.

2. Methodology

The authors developed a stochastic macro-financial model that merges deterministic macroeconomic debt-risk models with Kelly's gambling model (originally from information theory and evolutionary biology).

Core Model Mechanics

• Stochastic Capital Dynamics: The model treats an economy (or investor) as a capital $C_\tau$ evolving stochastically with an intrinsic growth rate $\gamma_\tau$.
• Debt and Leverage: The entity borrows money at an interest rate $\rho$. Leverage ($L$) is defined as the ratio of total assets to equity.
• Two Strategies:
  • Strategy A (One-time borrowing): Analogous to a mortgage. The investor borrows once, and the debt grows with interest until repayment. Bankruptcy occurs if debt exceeds capital at repayment.
  • Strategy B (Periodic borrowing): Analogous to state or corporate budgeting. The investor borrows continuously (periodically) to maintain growth. This is the primary strategy used for macroeconomic calibration.
• Solvency Condition: Solvency is defined not just by the existence of debt, but by the ability to repay debt at a specific time $t_p$ without the capital becoming negative.
• Carbon Linkage: The model links economic wealth ($C_\tau$) to cumulative emissions ($E$) via the Kaya identity:
  $$\epsilon(\tau, C_\tau) = C_\tau \times I(\tau)$$
  Where $I(\tau)$ is the carbon intensity (emissions per unit of GDP), which decreases over time due to decarbonization efforts.
• Path-Dependent Intensity: The authors introduce a metric $I_\tau(\{C_\tau\})$ to measure the cumulative carbon cost of a growth trajectory, which accounts for the history of emissions rather than just the current annual rate.

Data Calibration

The model was calibrated using multi-decade data (1998–2022) for four distinct economies: USA, China, France, and Denmark. Key parameters extracted included:

• Average leverage ($L$)
• Intrinsic growth rate ($\gamma$)
• Interest rates ($\rho$)
• Carbon intensity trends ($I(\tau)$)

3. Key Contributions

1. Novel Modeling Framework: This is the first model to explicitly link fiscal leverage with cumulative CO2 emissions using a stochastic credit-risk framework calibrated to real-world macroeconomic data. It moves beyond the exogenous growth assumptions of standard IAMs.
2. The "Double Debt" Concept: The paper conceptualizes debt as creating a "double debt": a financial debt (obligation to repay with interest) and a climatic debt (consumption of future energy/carbon budget).
3. Identification of the Leverage Frontier: The model identifies a critical threshold where the intrinsic growth rate falls below the effective interest rate. Beyond this point, additional credit primarily generates systemic risk (bankruptcy) rather than sustainable growth.
4. Quantification of Carbon Lock-in: The study provides a mathematical explanation for why cumulative emissions remain proportional to cumulative GDP, even in countries with aggressive decarbonization policies.

4. Key Results

A. The Carbon Lock-In Effect

• Proportionality: The model demonstrates a robust coupling between cumulative GDP and cumulative emissions. Even when annual carbon intensity ($I(\tau)$) declines, the path-dependent intensity ($I_\tau$) remains high because early, carbon-intensive years of growth dominate the cumulative total.
• Amplification by Leverage: Higher leverage ($L$) boosts GDP growth, which in turn increases cumulative emissions. The study finds that a few percentage points increase in leverage can lead to an order-of-magnitude increase in cumulative emissions per capita over 25 years.
• Ineffectiveness of Efficiency Alone: Improvements in energy efficiency (lowering $I(\tau)$) are often offset by the increased scale of economic activity driven by debt.

B. Solvency and Financial Instability

• The Solvency Threshold: The probability of long-term solvency approaches zero when the intrinsic growth rate ($\langle \gamma \rangle$) is less than the interest rate ($\rho$).
• Strategy B Risk: Continuous borrowing (Strategy B) carries higher default risk than one-time borrowing (Strategy A) for the same leverage level, as it relies heavily on sustained growth to service debt.
• The Trade-off: High leverage strategies maximize short-term GDP growth but drastically increase the risk of default and cumulative emissions. The "optimal" leverage for growth (ignoring climate costs) is often beyond the point of financial stability.

C. Empirical Validation

• Model Accuracy: The model accurately reproduces historical cumulative CO2 emissions per capita for the US, China, France, and Denmark without fitting to emissions data directly. It derives emissions from independently calibrated economic parameters (deficits, GDP, debt).
• Country Specifics:
  • USA: Heavily dependent on deficit spending (leverage $L \approx 1.06$ in 2023). Intrinsic growth is low (~1%), meaning most growth is credit-driven.
  • China: High leverage drives rapid growth, but the centralized credit allocation allows for some temporary decoupling by directing funds to green infrastructure, though cumulative emissions remain high.
  • Denmark: Runs fiscal surpluses (negative leverage). While territorial emissions decreased, consumption-based emissions remained stable due to "carbon leakage" (outsourcing emissions), highlighting that fiscal restraint alone does not guarantee decoupling without addressing trade.

5. Significance and Implications

• Challenge to Decoupling: The results challenge the feasibility of "green growth" (decoupling GDP growth from emissions) under prevailing credit-driven regimes. As long as growth relies on debt, cumulative emissions will rise.
• Policy Recommendations:
  • Credit Allocation: Achieving net-zero targets requires aligning credit allocation specifically with decarbonization objectives, not just general growth.
  • Rethinking Growth Targets: The paper suggests that fiscal stimulus policies based on debt-financed growth must be assessed for their climate implications and the risk of triggering financial instability.
  • Systemic Risk: There is a "leverage frontier" beyond which additional credit creates more risk (financial default + climate overshoot) than value.
• Theoretical Shift: The paper bridges the gap between ecological macroeconomics and financial theory, arguing that the "limits to growth" are not just physical (resources) but also financial (solvency constraints).

In conclusion, the paper argues that debt is a structural driver of the carbon crisis. It posits that without a fundamental shift in how credit is allocated and a decoupling of prosperity from the credit lever, economies will remain locked into high-carbon trajectories, risking both financial collapse and climate tipping points.