Emissions-Robust Portfolios

Khizar Qureshi, H. Oliver Gao · 2026

Read Emissions-Robust Portfolios by Khizar Qureshi, H. Oliver Gao (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-06507.txt.

Introduction

A fund can cut the smoke its firms make, but smoke data is full of slips. Khizar Qureshi and 1 co-author build funds that stand up to those slips in a 2026 paper. Their key tool trims each stock's gain by a smooth rule tied to how much smoke it makes per coin earned.

Four plain terms force that trim tool to one closed form; no other shape fits. Paired with doubt sets that hedge the slips, it gives fund plans that can be solved with care. Their dual terms can be read as prices on risk and smoke. In a big-cap US test with trades each month, the plan cut the smoke rate of its stocks by a vast share next to a flat-weight fund. Careful tests found no clear loss in gains or in the gain-to-risk rate.

Catalog record

CREATORKhizar Qureshi, H. Oliver Gao
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE111 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-06507

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Emissions-Robust Portfolios Khizar Qureshi, H. Oliver Gao Abstract We study portfolio choice when firm-level emissions intensities are measured with error. We introduce a scope-specific penalty operator that rescales asset payoffs as a smooth function of revenue-normalized emissions intensity. Under payoff homogeneity, unit-scale invariance, mixture linearity, and a curvature semigroup axiom, the operator is unique and has the closed form P j ( m ) ​ ( r , λ ) = ( 1 − λ / λ max , j ) m ​ r P^{(m)}_{j}(r,\lambda)=\bigl(1-\lambda/\lambda_{\max,j}\bigr)^{m}r . Combining this operator with norm- and moment-constrained ambiguity sets yields robust mean–variance and CVaR programs with exact linear and second-order cone reformulations and economically interpretable dual variables. In a U.S. large-cap equity universe with monthly rebalancing and uniform transaction costs, the resulting strategy reduces average Scope 1 emissions intensity by roughly 92% relative to equal weight while exhibiting no statistically detectable reduction in the Sharpe ratio under block-bootstrap inference and no statistically detectable change in average returns under HAC inference. We report the return–emissions Pareto frontier, sensitivity to robustness and turnover constraints, and uncertainty propagation from multiple imputation of emissions disclosures. Keywords robust optimization, portfolio selection, sustainable finance, carbon emissions, conic optimization 1 Introduction Climate change is a canonical global stock externality: the damage-relevant state variable is the atmospheric concentration of greenhouse gases, not the contemporaneous flow of emissions. Anthropogenic CO 2 concentrations exceeded 420 ppm in 2023, relative to a pre-industrial baseline of roughly 280 ppm ( 22 ) . A back-of-the-envelope calculation links temperature targets to the residual global carbon budget. Let B_{\max} denote the cumulative CO 2 emissions compatible with a maximum admissible warming \Delta T_{\max} , and let \kappa be the temperature response per gigaton of CO 2 . Then \displaystyle B_{\max} \displaystyle=\frac{\Delta T_{\max}}{\kappa} (1) \displaystyle\approx\frac{1.5^{\circ}\mathrm{C}}{4.8\times 10^{-4}\,{}^{\circ}\mathrm{C}/\mathrm{GtCO}_{2}} \displaystyle\approx 310~\mathrm{GtCO}_{2}. At current emissions rates of roughly 40\,\mathrm{GtCO}_{2}/\text{year} , this residual budget would be exhausted within a decade. (a) Net atmospheric addition of CO 2 (GtCO 2 /yr), 1959–2024, with a 95% confidence band. The band propagates NOAA Mauna Loa growth-rate uncertainty ( \pm 1.96\times 0.11 ppm) into mass units (1 ppm \approx 7.77 GtCO 2 ). (b) Stylized 1.5 ∘ C remaining budget, B_{0}=310 GtCO 2 , under alternative usage paths: constant 40 GtCO 2 /yr, linear decline to net-zero by 2032, exponential 5%/yr reduction. And late action (flat to 2028 then 15%/yr decline). Exhaustion years are annotated where applicable. Figure 1: From atmospheric accumulation to policy-relevant budgets. Panel (a) translates ppm growth into GtCO 2 /yr and shows measurement uncertainty, panel (b) illustrates how non-linear emissions pathways induce different curvature in the remaining-budget trajectory. Values are illustrative and sensitive to methodological choices (e.g., non-CO 2 forcers and probability thresholds). These numbers underscore a simple point: anthropogenic warming is driven by the cumulative stock of CO 2 . The IPCC Sixth Assessment Report (AR6) formalizes the remaining carbon quotas consistent with 1.5^{\circ}\mathrm{C} and 2^{\circ}\mathrm{C} pathways and emphasizes the need for deep, near-term mitigation 22 . Flow estimates from the Global Carbon Budget ( 16 ) document record or near-record fossil emissions, highlighting the gap between physical constraints and realized trajectories. The Paris Agreement’s Article 2.1(c) explicitly commits signatories to “making finance flows consistent with a pathway towards low greenhouse gas emissions and climate-resilient development” ( 31 ) . Complementary analyses by the UNEP Emissions Gap Report ( 32 ) quantify the shortfall between pledged policies and emissions pathways compatible with 1.5^{\circ}\mathrm{C} . Together, these developments motivate a quantitative finance response: design portfolio frameworks that deliver verifiable emissions reductions without sacrificing financial performance or risk discipline. Institutional investors sit at the transmission channel between these climate constraints and the real economy. Global capital markets intermediate on the order of $400 trillion in financial assets ( 7 ) . Redirecting even a modest fraction of this capital toward climate-compatible firms could generate first-order mitigation effects. Two frictions, however, impede such realignment. First, corporate greenhouse gas (GHG) disclosures are incomplete, noisy, and subject to strategic reporting and methodological disagreement across data vendors ( 5 ) . Second, asset managers are still evaluated through classical lenses—risk-adjusted return, drawdown control, and turnover. Credible portfolio design must therefore integrate two classes of uncertainty on equal footing: (i) estimation risk in corporate GHG disclosures and (ii) the familiar mean–variance trade-off governing financial optimization ( 23 ) . In § 4 we make this comparison concrete by benchmarking EAPO against deterministic inverse-intensity tilts that treat estimated intensities as known. (a) Global fossil and land-use CO 2 emissions by source, 1960–2023. (b) Decomposition of 2023 CO 2 emissions by source for major emitters. Figure 2: Global and 2023 CO 2 emissions by source. Source: Our World in Data, “CO 2 and Greenhouse Gas Emissions” dataset. Literature and conceptual foundations. Debate on the financial performance of ESG-tilted portfolios dates back to early comparisons of socially responsible and conventional mutual funds ( 28 ; 14 ) . Meta-analyses such as 15 find on average a weakly positive relation between ESG characteristics and financial performance. Production-based asset pricing models attribute ESG premia to profitability and investment channels ( 2 ) . Building on these findings, a growing literature studies the optimal construction of ESG-integrated and low-carbon portfolios. On the portfolio-design side, early low-carbon strategies imposed exogenous carbon-budget constraints on Markowitz frontiers or introduced decarbonization factors in index construction ( 1 ; 18 ) . Subsequent work incorporated climate Value-at-Risk measures ( 33 ) and climate-hedging overlays for liquid instruments ( 12 ) . Robust optimization provides the natural mathematical foundation for such extensions. Originating with the ellipsoidal uncertainty sets of 4 and the budget-of-uncertainty model of 6 , robust methods have been adapted for climate stress testing and transition risk analysis ( 7 ) . Yet, most existing approaches treat emissions exposures as known constants and either enforce hard constraints or apply ad hoc tilts, rather than modeling emissions as a noisy state variable and incorporating disclosure uncertainty within the optimization problem itself. Parallel work in macroeconomics and asset pricing documents that climate risk is economically material. Global output follows a nonlinear, concave temperature–productivity relation with large welfare losses at high temperatures ( 10 ; 19 ) . In asset markets, long-run temperature risk carries priced premia ( 3 ) , and damages are heterogeneous across sectors and geographies ( 11 ) . Transition risk now appears as a priced cross-sectional factor: carbon-intensive issuers exhibit higher realized returns consistent with exposure to a transition premium ( 8 ) , while option-implied volatilities reveal elevated compensation for downside climate risk ( 20 ) . Dynamic climate-hedging strategies reduce exposure to climate news using liquid equity instruments ( 12 ) . The ESG-efficient frontier formalizes the trade-off between risk-adjusted return and sustainability exposure ( 27 ) , and equilibrium analyses show that heterogeneous investor preferences can generate structural return differentials between “green” and “brown” assets ( 26 ) . Finally, 17 provide a comprehensive review and emphasize that climate-related risks are economically material while identification and measurement remain central challenges for empirical asset pricing. Regulatory and disclosure frameworks have begun to codify these practices. The EU’s Paris-Aligned Benchmark (PAB) and Climate Transition Benchmark (CTB) frameworks impose minimum decarbonization rates and transparency requirements ( 13 )

Read the complete text