Behind on Which Front? Knowledge, Creation and Deployment in Europe’s Position Relative to the United States and China in Business, Artificial Intelligence and Renewable Energy

Technical Summary: Behind on Which Front? Knowledge, Creation and Deployment in Europe's Position Relative to the United States and China

Problem Statement
The debate regarding whether the European Union (EU) has fallen behind the United States and China is central to European economic policy, yet it often conflates distinct stages of technological competitiveness into a single verdict. Existing literature typically isolates specific dimensions—such as patent counts, digitalization indicators, or trade flows—leading to contradictory conclusions about the same economies. This study addresses the lack of a unified framework that distinguishes between the production of knowledge, the conversion of that knowledge into commercial assets, and the deployment of technology. The core problem is determining not just if Europe is lagging, but where along the value chain the gap exists and how it has evolved relative to its two primary rivals.

Methodology
The study employs a three-pillar framework to measure the EU-27's position relative to the US and China across three domains: Business, Artificial Intelligence (AI), and Renewable Energy.

• Pillars:
  1. Knowledge: Scientific output (articles, publications).
  2. Creation: The conversion of knowledge into commercial assets (corporate R&D, venture capital, frontier models, patents, computing capacity, and exportable manufacturing).
  3. Deployment: The adoption and diffusion of technology (renewable electricity share, per capita generation, AI talent concentration).
• Data: The analysis utilizes 14 harmonized indicators from six public sources (World Bank, JRC Scoreboard, Stanford HAI, Our World in Data, Eurostat Comext, UN Comtrade) covering the period 2005–2024.
• Statistical Approach:
  • Indicators are expressed as log ratios ($r_{i,t} = \ln(X_{EU}/X_{Comparator})$), allowing for symmetric interpretation and direct calculation of annual percentage changes.
  • Trends are estimated using Ordinary Least Squares (OLS) with HAC-robust standard errors (Newey–West) to account for heteroskedasticity and autocorrelation.
  • A pre-specified structural break test (2020) is applied to detect trajectory changes post-pandemic and post-AI boom.
  • Indices: Geometric means of log ratios are calculated for each pillar. "Leave-one-out" tests assess the sensitivity of these indices to individual indicator composition.
  • Robustness: Checks include an "Extended Europe" aggregate (EU-27 + UK, Switzerland, Norway), long-window analysis (2010–2023), and cross-validation of trade data.

Key Results
The analysis reveals a significant asymmetry in Europe's competitive position, particularly when compared to the United States, and a broad decline relative to China.

• Relative to the United States (2023):

  • Knowledge: The EU leads, with a ratio of 1.34 (producing ~34% more science).
  • Deployment: The EU leads, with a ratio of 1.23.
  • Creation: The EU lags significantly, with a ratio of 0.33. The EU generates only about one-third of the American level of corporate R&D, AI capital, frontier models, and clean-tech exports combined.
  • Trend: The asymmetry is stable; the creation index has not improved since 2017. The gap is most severe in ICT corporate R&D (0.08 ratio) and notable AI models (0.10 ratio).

• Relative to China (2023):

  • Knowledge: The EU has lost its advantage, dropping from parity (1.07 in 2017) to 0.55 (a 10.4% annual decline).
  • Creation: The EU has lost its advantage, dropping from 0.62 to 0.29 (a 12.5% annual decline).
  • Deployment: The EU retains a lead (1.52), though this is narrowing (4.3% annual decline).
  • Trend: All three pillars show significant negative trends against China. The EU's share of R&D among the world's top 1,000 corporate investors fell from 24.7% (FY2010) to 16.4% (FY2024).

• Sectoral Nuances:

  • AI: The gap with the US is concentrated in capital, frontier models, and patents, while research output and talent concentration remain comparable.
  • Renewables: The EU leads the US in the share of renewable electricity but has been overtaken in per capita wind and solar generation. In clean-technology manufacturing (wind turbines, batteries), the EU leads the US but trails China drastically (exports fell from 0.79 to 0.17 of China's level).

Key Contributions

1. Decomposition of Competitiveness: The paper provides the first empirical evidence separating the "knowledge," "creation," and "deployment" stages for the EU, US, and China simultaneously across business, AI, and energy.
2. Identification of the "Conversion" Gap: The study demonstrates that Europe's primary deficit is not in science or adoption, but in the intermediate stage of converting knowledge into firms, capital, and scalable products.
3. Reconciliation of Literature: It explains why previous studies reached opposite conclusions: those focusing on patents or scientific output often find Europe strong, while those focusing on commercial assets find it weak. The framework shows these are measuring different stages of the same process.
4. China as the Primary Competitor in Manufacturing: The data challenges the view that the US is Europe's main competitor in green manufacturing; the evidence indicates China has overtaken Europe in both knowledge and manufacturing scale.

Significance and Claims
The paper claims that Europe has indeed fallen behind, but the nature of the deficit is specific. The "verdict" of falling behind is driven almost entirely by the creation pillar.

• Policy Implication: The study argues that policy should target the conversion stage (venture capital, firm scaling, sectoral mix of research) rather than solely increasing public research spending or adoption incentives.
• Fragility of Deployment: The EU's perceived lead in renewable energy is shown to be fragile, resting heavily on the share of renewables in the grid (aided by hydropower) rather than the per-capita build-out of new capacity, which is critical for an AI-driven economy.
• Strategic Focus: The author concludes that a strategy relying on Europe's strong science base is insufficient against China, where both the knowledge base and conversion capabilities have advanced. The framework suggests that competitiveness should be judged by the weakest stage (creation) rather than averages, as a strong knowledge base cannot compensate for a failure to commercialize.

The study explicitly avoids claiming that regulation is the sole cause of the gap, noting that while AI research and skills are comparable to the US, the lack of capital and scale suggests finance and industrial structure are the binding constraints. The findings are presented as a diagnostic tool to inform the implementation of recent European competitiveness reports (e.g., Draghi, 2024).