Carbon Footprints of the Equity Portfolios of Chinese Fund Firms

Global pledges under the Paris Agreement appear insufficient to limit warming to 1.5 °C, placing pressure on the financial sector to align capital flows with net-zero targets. China, the world’s largest CO₂ emitter, has pledged to peak carbon emissions by 2030 and achieve carbon neutrality by 2060. ESG investing is growing in China, but the lack of unified, compulsory disclosure of investment carbon footprints poses challenges and risks for financial institutions. International frameworks such as PCAF and TCFD provide standardized guidance to measure and disclose financed emissions, yet Chinese institutions face data limitations, especially regarding firm-level emissions and Scope 3 accounting. This study focuses on Scope 1 and Scope 2 emissions and applies TCFD-aligned methods to analyze financed emissions and carbon intensities for equity portfolios of 105 Chinese fund firms (2010–2020), identify drivers of financed emissions, and explore decarbonization potential to inform investors and policymakers.

The paper situates its analysis within the emerging standards for investment carbon accounting developed by PCAF and TCFD, which enable consistent measurement and disclosure of financed emissions. Prior literature highlights regional disparities in firm-level emissions disclosure—more prevalent in Europe and North America, often mandated (e.g., UK)—and limited in developing countries, including China. The importance and difficulty of incorporating Scope 3 emissions into portfolio carbon footprints are emphasized, given supply-chain complexities and data scarcity. Studies show foreign asset managers have generally reduced financed emissions in alignment with climate goals, while Chinese managers’ emissions trends relate to investment structures with heavy exposure to carbon-intensive sectors. The paper also references work on performance and uptake of ESG in Chinese mutual funds and the impact of disclosure mandates on emissions and financial performance.

Scope and metrics: The study analyzes equity portfolios of 105 Chinese public fund firms from 2010 to 2020, focusing on investee companies’ Scope 1 and Scope 2 emissions due to data limitations for Scope 3. Financed emissions and two carbon intensity metrics—Weighted Average Carbon Intensity (WACI) and Carbon Emissions to Revenue Intensity (CERI)—are computed following TCFD guidance. Financed emissions Ft = Σi (Iit/Mt) * Eit, where Iit is the market value of holding i and Mt is the total portfolio market value. WACI = Σi (Iit/Mt) * (Eit/Rit). CERI = (1/Mt) * Σi (Iit * Eit/Rit). Firm-level emissions estimation: When firm-level Scope 1 and 2 data are unavailable, emissions Eijt are estimated using an average-data approach with environmentally-extended input-output (EEIO) data: Eijt = Rijt * Cjt / Ojt, where Rijt is company revenue, Cjt is sectoral emissions, and Ojt is sectoral output for sector j in year t. Data sources: Economic output from GTAP (2011, 2014, 2017; used to fill adjacent years); Scope 1 from CEADs (42 sectors); Scope 2 from GTAP-E (electricity consumption); equity holdings, market values, revenues, and SIC codes from S&P Capital IQ Pro. The analysis aggregates sectors to 28 categories and distinguishes carbon-intensive vs non-intensive sectors. The sample includes firms whose equity portfolios are covered by S&P Capital IQ Pro; most investees are Chinese-listed companies headquartered in China. Data processing: To address missing intermediate years, the study carries forward/backward the closest GTAP/GTAP-E data (e.g., 2011 values used for 2010, 2012–2013). Investments are attributed proportionally by portfolio weight. Sector mapping is based on SIC four-digit codes to the 28-sector aggregation. Validation: Estimated emissions are compared against self-reported data for 56 high-emitting listed companies (electricity, cement, steel). Discrepancies are quantified and uncertainty ranges explored, noting heterogeneity in sub-sectors (e.g., renewable vs thermal power) as a source of divergence.

• Total financed emissions of Chinese fund firms’ equity portfolios fluctuated 2010–2015 (100.26 Mt in 2010, peak 183.11 Mt in 2012, falling to 87.21 Mt in 2015), then rose steadily to a second peak of 168.40 Mt in 2018 and stabilized at 159.00 Mt in 2020.
• Carbon-intensive sectors dominated financed emissions; Electricity, Gas and Water consistently contributed about 50% of total financed emissions and largely drove the 2012 peak before declining to 42.80 Mt in 2015. Metals Smelting also showed significant variation (e.g., down to 9.75 Mt in 2015).
• In 2020, the Top 10 fund firms accounted for 88.77 Mt, or 55.83% of total financed emissions. Individual firms showed diverse trajectories; e.g., Fund 8’s financed emissions increased by 3.10× from 2010 to 2020; Fund 5 by 2.98×; Fund 1 by 0.98×. Electricity, Gas and Water were primary drivers for several top firms (e.g., Fund 1: 9.57 Mt in 2020, 72.77% of its total; Fund 5: 0.08 Mt in 2010 to 5.62 Mt in 2020). Metals Smelting notably contributed to Fund 2, Fund 8, and Fund 9 by 2020.
• WACI displayed a generally downward trend with peaks at 624.76 tCO₂/M$ in 2012 and 577.45 tCO₂/M$ in 2018. Top 10 firms had slightly higher average WACI but declined over time; e.g., Fund 4 from 635.40 (2010) to 282.25 (2020), linked to reduced allocations to Coal Mining (-4.66%) and Electricity, Gas and Water (-0.80%). Fund 10 declined from 685.14 to 361.29 tCO₂/M$.
• CERI exhibited a bimodal pattern: 356.49 (2010) rising to 654.36 (2012), falling to 415.07 (2015), then peaking at 608.06 (2018) before declining to 493.42 (2020). For the Top 10 firms, average CERI trended upward since 2016, reaching 673.86 in 2020.
• Revenue exposure to carbon-intensive sectors was generally low and declined among Top 10 firms. Example: Fund 5’s revenue share from carbon-intensive sectors fell from 33.51% (2010) to 12.08% (2020), driven by reductions in Coal Mining (-11.31%) and Metals Smelting (-14.71%). Petroleum Processing and Coking, Nonmetal Products, and Electricity, Gas and Water underperformed in revenue due to shrinking demand and decarbonization.
• Decarbonization profiles: Three categories identified (2016–2020). Fund a: rising financed emissions and rising WACI/CERI (22.05% and 74.70% increases), heavy exposure to Metals Smelting and Electricity, Gas and Water (~70% since 2018). Fund b: declining financed emissions and declining WACI (474.98→163.38 tCO₂/M$) and CERI (576.34→224.22 tCO₂/M$), with major divestments from carbon-intensive sectors and reallocations to high-tech (electronics/telecoms; electric equipment). Fund c: rising financed emissions but declining WACI (-63.31%) and CERI (-35.84%), with shifts towards electric equipment and electronics and reduced exposure to Petroleum Processing and Coking and Electricity, Gas and Water.
• Validation against self-reported data for 56 high-emitting firms (2021) showed sectoral discrepancies: electricity +22.21%, cement -85.82%, iron -34.21% (estimated vs reported), with wide uncertainty ranges, especially in electricity due to renewable vs thermal mix heterogeneity.

The analysis demonstrates that Chinese fund firms’ financed emissions have increased since 2015, and portfolios remain heavily exposed to carbon-intensive sectors, implying elevated carbon-related risks. While WACI has generally declined, indicating some reduction in exposure, intensity improvements have not fully offset rising absolute financed emissions, particularly among leading firms that contribute a large share of total financed emissions. The findings underscore the need for Chinese institutional investors to align with international best practices by setting phased decarbonization targets, enhancing carbon disclosure, and improving portfolio construction to reduce exposure to high-emitting sectors. Strategic reallocation toward high-tech and energy-efficient sectors (e.g., electronics, electric equipment) and renewable energy can reduce financed emissions and carbon intensity while maintaining economic performance. The typology of fund behaviors (rising/rising; falling/falling; rising/declining) provides a roadmap for targeted interventions: high-risk funds should divest more aggressively from carbon-intensive assets; leaders should continue to strengthen disclosure and emissions reduction commitments; and transition funds should accelerate shifts towards lower-carbon holdings to consolidate intensity gains into absolute emissions reductions.

The study applies TCFD-aligned metrics to quantify financed emissions and carbon intensities for the equity portfolios of 105 Chinese fund firms over 2010–2020, revealing rising financed emissions since 2015, dominance of carbon-intensive sectors (notably Electricity, Gas and Water), and substantial contributions from the largest firms. Although portfolio carbon intensities (WACI, CERI) show improvement in some cases, absolute financed emissions remain elevated. The authors recommend that Chinese institutional investors: (1) adopt clear, phased portfolio decarbonization targets; (2) enhance and standardize carbon disclosure practices; (3) reallocate capital away from carbon-intensive sectors toward high-tech and renewable sectors; and (4) integrate robust ESG scoring and stewardship to drive emissions reductions among investees. Future research should incorporate more accurate firm-level emissions, expand to Scope 3 with improved attribution methods, refine sectoral heterogeneity (e.g., separating renewable vs thermal power), and further validate estimates against disclosures to reduce uncertainty.

• Scope coverage: Only Scope 1 and Scope 2 emissions are included; Scope 3 is excluded despite its importance, limiting the completeness of portfolio carbon risk assessment.
• Emissions estimation: Heavy reliance on an industry-average EEIO method (E = revenue × sectoral emissions/output) introduces uncertainty and may not reflect firm-level heterogeneity (e.g., renewable vs thermal power generation within the electricity sector).
• Data availability: Very few Chinese listed companies disclose emissions (e.g., 211 in 2020; ~5.1% of all listed), constraining validation and calibration.
• Temporal resolution: GTAP/GTAP-E data are only available for select years; interpolation by carry-forward/backfill can introduce bias.
• Potential double counting and attribution: Investment emissions attribution may double count across value chains; Scope 3 attribution rules are not implemented (100% attribution used for Scope 1–2 per PCAF), which may misstate responsibility across portfolios.
• Validation discrepancies: Notable gaps between estimated and self-reported emissions in electricity, cement, and iron sectors, with wide uncertainty ranges due to sub-sector differences and corporate transition strategies.