Green technology emerges as China’s strongest lever against carbon intensity

China’s battle against carbon emissions has entered a new phase in which the weapons are no longer only smokestack scrubbers and fuel-economy rules, but algorithms, mobile payment networks, data centers and clean-energy patents. A study published in Discover Sustainability by Anam Tariq and colleagues examines how four pillars of modern technological change—financial technology (FinTech), digitalization, information and communication technology (ICT), and green technology (Gtech)—shape China’s carbon intensity, defined as carbon dioxide emissions per unit of economic output. Using annual data from 2000 to 2022, the researchers reach a conclusion that challenges conventional econometric practice: the impact of technology on emissions is not a single number, but a shifting landscape that changes with emission conditions and with time.

The core problem the study addresses is one of averages. Most previous analyses of technology and pollution rely on average-effect estimates, which assume that a variable’s influence on carbon intensity is uniform whether emissions are at their lowest or their highest. The authors argue that this assumption is untenable. FinTech, digitalization, ICT and green technology do not act homogeneously; their environmental consequences can differ significantly depending on the level of carbon intensity in a given period. An economy experiencing unusually high emissions may respond to digital investment very differently from one operating near its cleanest historical performance, and an average-effect model would blur precisely the distinctions policymakers need to understand.

To capture this heterogeneity, the team deployed an unusually rich methodological arsenal. Their central technique is Quantile on Quantile Regression (QQR), which estimates how different quantiles of each explanatory variable interact with different quantiles of carbon intensity, effectively mapping the relationship across the full joint distribution rather than at a single average point. They complement this with Multivariate Quantile on Quantile Regression (MQQR), which assesses the combined effect of the technological ecosystem as a whole, and with two time-frequency tools borrowed from signal processing: Wavelet Coherence (WC) and Wavelet Quantile Correlation (WQC). These wavelet methods allow the researchers to track how strongly the variables move together across short-, medium- and long-run periods throughout the 23-year sample, revealing dynamics that a fixed-coefficient regression would miss entirely.

The results are striking in their asymmetry. At lower quantiles of carbon intensity—periods when emissions were relatively low—FinTech exhibits weaker and statistically insignificant effects on carbon intensity. In other words, digital finance does not appear to be a reliable emissions-reduction tool when the economy is already running clean. This finding complicates the popular narrative that fintech innovation automatically greets environmental goals; its leverage appears conditional rather than constant. For regulators, the implication is that promoting FinTech as a climate instrument requires attention to the prevailing emission regime, not just to the volume of digital financial activity.

Digitalization tells a clearer story. Its effect on carbon intensity is mostly negative, and most pronounced at high emission quantiles—precisely where reductions matter most for climate outcomes. When China’s carbon intensity was at its dirtiest, the expansion of digital processes, automation and data-driven efficiency delivered some of its strongest curbing effects. This makes intuitive technical sense: digitalization substitutes information flows for material ones, optimizes logistics, reduces transaction frictions and enables real-time monitoring of energy use, and these mechanisms bite hardest when inefficiency and emissions are at their peak. Digitalization, the study suggests, is a technology whose climate dividend scales with the size of the problem it confronts.

ICT proves the wildcard. Unlike the other variables, information and communication technology exhibits a nonlinear and more complex pattern across quantiles of carbon intensity. Its effect flips and shifts in strength depending on where in the distribution the economy sits, reflecting what environmental economists have long described as a double-edged dynamic. ICT can compress carbon intensity by enabling dematerialization and smarter infrastructure, yet it also expands the electricity-hungry footprint of networks, servers and devices. The study’s quantile-based evidence shows that neither a uniformly optimistic nor a uniformly pessimistic reading of ICT is defensible; its environmental role depends on the emission context in which it operates.

The clearest winner is green technology. Across almost all quantiles of carbon intensity, Gtech shows the strongest and most stable negative effect of any explanatory variable in the model. Green technology patents and innovations—covering renewable energy, energy efficiency and pollution control—consistently reduce emissions per unit of output regardless of whether the economy is running clean or dirty. This stability is the crucial technical distinction: while FinTech’s effect fades at low emission levels and ICT’s effect oscillates, green technology behaves like a dependable structural reducer of carbon intensity. For policymakers deciding where to concentrate research funding and industrial policy, that consistency is a powerful argument.

Perhaps the most consequential finding comes from the multivariate analysis. The MQQR results reveal that the overall effect of the technological ecosystem—FinTech, digitalization, ICT and green technology acting together—is more emission-reducing than the sum of the individual impacts estimated by conventional QQR. The technologies reinforce one another: digital infrastructure amplifies the diffusion of green patents, fintech channels capital into clean innovation, and ICT enables the monitoring and optimization that make both effective. This complementarity means that treating these tools as separate policy levers understates their potential. An integrated technological strategy, in which digital and financial innovation are designed to accelerate green technology deployment, delivers a greater climate payoff than the same investments made in isolation.

The wavelet analysis adds the temporal dimension that annual averages conceal. By examining coherence across short-, medium- and long-run periods for the full sample, the study shows that the relationships among these technologies and carbon intensity are not static over the two decades examined. China’s economic transformation—from WTO accession in 2001 through its industrial supercycle to its dual-circulation digital era—means that the technology-emission nexus itself evolved, with structural breaks detected in the data confirming regime changes along the way. Time-frequency methods of this kind are increasingly seen as essential in energy economics, because policy prescriptions drawn from one era may simply not transfer to another.

The study’s implications align explicitly with the United Nations Sustainable Development Goals, particularly SDG 7 on affordable and clean energy, SDG 9 on industry, innovation and infrastructure, SDG 12 on responsible consumption and production, and SDG 13 on climate action. For a country that produces roughly a third of global carbon emissions while simultaneously leading the world in renewable energy capacity, electric vehicle manufacturing and digital payments, understanding which technologies actually bend the carbon-intensity curve—and under what conditions—is a question of global significance. The authors’ evidence underscores the significance of FinTech, digitalization, ICT and green technology in supporting the transition to energy-efficient and low-carbon economic systems in China. The broader lesson for the international community is methodological as much as substantive: in a world of nonlinear, state-dependent environmental dynamics, the era of single-coefficient answers is over, and quantile-aware, time-aware analysis is becoming the standard against which credible climate-economics research will be judged.

Subject of Research: The asymmetric effects of FinTech, digitalization, ICT and green technology on carbon intensity in China.

Article Title: Decoding carbon intensity in China through FinTech, digitalization, and green technology

Article References: Tariq, A., Khan, N., Shah, S. Z. A., & Rizwan, M. F. (2026). Decoding carbon intensity in China through FinTech, digitalization, and green technology. Discover Sustainability. https://doi.org/10.1007/s43621-026-04587-7

Image Credits: AI Generated

DOI: 10.1007/s43621-026-04587-7

Keywords: carbon intensity, China, FinTech, digitalization, ICT, green technology, quantile on quantile regression, wavelet coherence, climate policy, SDG 13, energy transition, emissions