General Equilibrium models, while useful for specific applications, are ill-suited to the complexities of climate policy. As the stakes continue to rise, the need for better modelling approaches becomes increasingly urgent. By embracing dynamic realism, we can develop models that provide more accurate and actionable insights, enabling policymakers to craft effective strategies for mitigating climate change.
Macroeconomic modelling plays a critical role in shaping climate policies by assessing the costs and benefits of various actions. However, reliance on General Equilibrium (GE) models – such as the Dynamic Integrated Climate-Economy (DICE) model – has come under scrutiny for failing to capture the complexities of climate change.
A recent working paper from the Institute of New Economic Thinking (INET), A New York City-based non-profit economic think tank, explores the inadequacies of traditional macroeconomic models in addressing climate change and the transition to a sustainable economy. The authors argue that standard models, like those based on Dynamic Stochastic General Equilibrium (DSGE), often assume smooth adjustments and market efficiency, which fail to capture the chaotic and non-linear nature of ecological and economic systems under environmental stress.
The Problem with General Equilibrium Models
GE models assume that markets reach a balance where supply equals demand, abstracting away from real-world complexities. While useful for theoretical analysis, their application to climate policy often oversimplifies critical dynamics, especially in three areas:
- Temporal Independence:These models assume that economic decisions in one period have no bearing on the next. For instance, the cost of cutting emissions in 2050 is modelled independently of past or future abatement efforts. This disregards the time, effort, and infrastructure investments required to transition to a low-carbon economy. Real-world systems involve path dependence, where earlier actions influence future costs and opportunities.
- Inertia and Transitional Costs: GE models often neglect the inertia of existing infrastructure and technology. Replacing or retrofitting power plants, transportation systems, and industrial facilities involves significant costs and delays. Historical evidence shows that transitions, such as adopting renewable energy or phasing out fossil fuels, take decades.
- Induced Learning and Innovation: The assumption of static abatement costs ignores how investments in research and development (R&D) can drive down costs over time. For example, advancements in renewable energy technologies have dramatically reduced costs, a trend absent in most equilibrium-based models.
These shortcomings lead to policy recommendations that may misrepresent the challenges and opportunities of climate action. For instance, static cost assumptions could undervalue early investments in clean technologies or misjudge the urgency of emissions reductions. This could result in underestimating the benefits of proactive climate policies and delaying critical actions.
Toward Dynamic Realism in Climate Modelling
To better address the challenges of the Anthropocene – a period defined by significant human impact on the Earth’s systems – the paper advocates for “real-world” modelling approaches. These include integrating biophysical constraints, feedback loops between the economy and environment, and acknowledging the risk of tipping points.
Alternative approaches emphasise dynamic realism, which accounts for the temporal and interconnected nature of economic and environmental systems. Key components include:
- Path Dependence: Recognising how past decisions shape future possibilities allows for better modelling of infrastructure transitions and cumulative emissions effects.
- Learning-by-Doing: Incorporating R&D and innovation into models shows how early investments can reduce long-term costs and accelerate the adoption of clean technologies.
- Sectoral and Regional Variability: Climate impacts and mitigation opportunities vary widely across sectors and regions. Dynamic models that reflect these nuances can provide more actionable insights for policymakers.
Developing more realistic models is not without challenges, however. They require greater data inputs, complex algorithms, and interdisciplinary collaboration. However, the potential benefits – more accurate assessments of climate policies and their socio-economic impacts – far outweigh these costs. Policymakers and researchers must advocate for models that reflect the dynamic interplay of economic, environmental, and technological systems to address the pressing realities of climate change effectively.
The authors of the INET working paper also emphasise a shift from growth-focused metrics toward models prioritising resilience, equity, and sustainability. Their framework challenges traditional assumptions of infinite substitutability between natural and manufactured capital and proposes a more grounded view of economic interactions within planetary limits.
This research underscores the urgency of reforming economic tools to ensure they are fit for managing the intertwined climate and economic crises, offering pathways for policymakers to incorporate ecological realities into decision-making processes.