Technology modularity shapes latecomer cost convergence in renewables

Technical Summary: Technology Modularity Shapes Latecomer Cost Convergence in Renewables

Problem Statement
The transition to net-zero electricity relies heavily on solar photovoltaics (PV) and onshore wind. While global average costs for these technologies have fallen rapidly, the affordability of the energy transition is determined by national installed costs, particularly in emerging economies. Existing technological diffusion theories often assume a "latecomer advantage," where late-adopting countries seamlessly inherit the experience and cost reductions of early movers. However, empirical evidence regarding whether this advantage holds for renewable energy remains mixed. A critical gap exists in understanding when, where, and under what technological conditions latecomers inherit global cost declines. Current energy-system models frequently rely on uniform global learning curves, potentially obscuring the significant heterogeneity in national cost trajectories and the specific constraints faced by developing nations.

Methodology
The authors introduce the Stage-Cost Assessment of Learning and Experience (SCALE) framework to reconstruct national installed-cost trajectories for utility-scale solar PV and onshore wind across more than 160 countries. The methodology proceeds through six integrated steps:

1. Data Harmonization and Imputation: Given sparse historical cost data (only 241 country-year observations for PV and 563 for wind from 2000–2022), the authors use a random-forest model to impute missing country-year costs based on techno-economic covariates (e.g., GDP, cumulative capacity, global component prices). These reconstructed series are smoothed using LOESS (locally estimated scatterplot smoothing) curves.
2. Take-off Year Identification: A three-tier reconciliation process identifies the year each country began sustained deployment, utilizing the Formative Phase Bayesian change-point Indicator (FOBI) and a complementary cascade-gated model with structural-break tests.
3. Cost-coupled Stage-Pathway Diffusion Model (C-SPDM): This model classifies national trajectories into six sequential deployment stages (from Formative Phase to Mature Penetration) based on annual new capacity, installed cost, cumulative capacity, and penetration. Countries are then grouped into four pathways: Front-Runners, Gradualists, Leapfroggers, and Laggards. Notably, for onshore wind, the Gradualist and Leapfrogger pathways are collapsed into a single "Late Mover" class due to weaker transferability.
4. Learning Rate Estimation: Using Wright's Law, the authors calculate national learning rates (the fractional cost reduction per doubling of cumulative capacity) for each country and pathway.
5. Probabilistic Projection: Future cost trajectories to 2050 are projected using Monte Carlo simulations based on historical learning-rate distributions under various scenarios.
6. Policy Scenario Analysis: Counterfactual scenarios (Business-as-Usual, Accelerated Deployment, Leapfrog Learning, Technology Transfer, and Combined Policy) are applied to "Laggard" economies to quantify the impact of interventions on cost convergence.

Key Results
The study reveals a profound asymmetry in cost convergence driven by technology architecture:

• Solar PV (Modular Architecture): Latecomer advantages are strong. Emerging-economy "Laggards" exhibit a median national learning rate of 27.5%, significantly higher than the 18.0% observed in developed-economy Laggards. This reversal is attributed to the modular, globally traded nature of PV modules, which allows learning to transfer across borders via supply chains. Latecomers enter at lower costs and diffuse roughly twice as fast as early movers.
• Onshore Wind (Site-Specific Architecture): Latecomer advantages are weak or non-existent. Emerging-economy Laggards have a median learning rate of only 9.7%, compared to 21.6% for developed counterparts. Wind deployment relies heavily on site-specific engineering, civil works, and grid connection, which limits the transferability of global turbine learning. Consequently, wind costs face an "enduring geographical floor," and late entrants often face costs similar to or higher than early movers.
• Decoupling of Growth and Cost: Rapid capacity growth does not automatically equate to domestic cost learning. For wind, capacity growth often peaks before cost parity is achieved, and many emerging economies remain in intermediate stages where deployment expands without significant cost convergence.
• Future Projections: Under Business-as-Usual scenarios, cost gaps persist through 2050. However, "Combined Policy" interventions (pairing deployment scale with technology upgrading) can effectively close the PV cost gap for emerging markets. In contrast, wind costs remain bound by country-specific institutional and geographic constraints, with variance decomposition showing that "Country Effect" explains over 56% of cost variance for wind, compared to 48% for PV.

Significance and Claims
The paper claims to provide empirical evidence that technology architecture governs the translation of global learning into local affordability. The authors argue that the assumption of a universal global learning curve is insufficient for energy-system modeling. Instead, models must distinguish between:

1. Globally transferable component learning (characteristic of modular technologies like PV), where latecomers can leverage global supply chains for rapid cost convergence.
2. Locally conditioned installed-cost learning (characteristic of site-specific technologies like wind), where cost reduction depends on domestic capabilities in project delivery, infrastructure, and regulation.

The significance of these findings lies in their implications for climate finance and policy. The authors assert that treating renewables as a homogeneous asset class is a strategic error. For solar PV, policies should focus on combining access to global learning with domestic market expansion. For onshore wind, technology transfer alone is insufficient; successful convergence requires sustained investment in domestic project-delivery capacity, grid planning, and industrial learning. The study concludes that recognizing this heterogeneity is essential for accurately modeling transition feasibility and financing strategies in emerging economies.