面對碳定價機制之全球擴張，內部碳定價（Internal Carbon Pricing, ICP）被視為企業管理氣候轉型風險之核心工具，惟其策略有效性長期缺乏前瞻性之實證評估。本研究建構結合機器學習與 SBM-DEA 之三階段前瞻性評估架構：以閘控循環單元（Gated Recurrent Unit, GRU）對美國上市企業之投入、期望產出與非期望產出進行逐年滾動之 walk-forward 預測（並以 LSTM、LightGBM 及線性迴歸基準為對照），將預測值導入 Tone (2004) SBM-Undesirable 模型，依序檢驗 ICP 採用有效性、工具型態有效性，以及未採用企業之碳價敏感度；分析期間為 2018 至 2023 年，CDP 配對樣本共 1,061 個 firm-years。

實證結果顯示：第一，ICP 採用企業之前瞻效率在高碳與低碳產業均顯著低於未採用企業（Mann-Whitney 檢定，p<0.0001 與 p<0.0001）；惟轉換者事件分析顯示，企業採用 ICP 前後之效率並無顯著變化，意味橫斷面之效率差距主要源於選擇效應——面臨較高轉型壓力之企業更傾向採用 ICP——而非採用行為本身之負面效果。第二，ICP 工具型態（金流型、非金流型與混合型）之間未呈現顯著之效率分化。第三，未採用企業之前瞻效率隨情境碳價（25 至 125 USD/tCO2e）單調下降且幅度重大，高碳與低碳產業之平均效率分別累計下降 23.1% 與 19.3%，企業層級之效率損失中位數高碳逾五成、低碳近五成（表 4-20）。上述發現於以 LightGBM 預測值重建之分析中方向完全一致。本研究據此將 ICP 重新定位為較側重於風險管理面向之工具（廣義之管理績效本即涵蓋風險管理，二者並非全然互斥）：其策略價值不在採用者之當期效率溢酬，而在其所對沖之碳價轉型風險之真實性與規模；配合導入近乎零成本之轉換者證據與「首段碳價調升衝擊最大」之型態，及早導入 ICP 即為低成本之理性風險管理決策。

As carbon pricing expands worldwide, internal carbon pricing (ICP) is a core tool for managing climate-transition risk, yet its effectiveness is rarely assessed on a forward-looking basis. Using U.S.-listed firms over 2018–2023, this study builds a three-stage framework integrating machine learning with slacks-based measure data envelopment analysis (SBM-DEA). A Gated Recurrent Unit network—benchmarked against LSTM, LightGBM, and linear models—produces walk-forward forecasts of inputs (selling and administrative expense, net property, plant and equipment, research and development), a desirable output (operating income), and an undesirable output (Scope 1 and 2 emissions); an SBM-Undesirable model measures forward-looking efficiency, and a carbon-weighted variant stress-tests non-adopters across prices of US$25–125 per tonne.

Prediction accuracy is high (up to 0.97; 0.653 and 0.862 for operating income). ICP adopters show significantly lower forward-looking efficiency in both high- and low-carbon industries (p < 0.0001); yet difference-in-differences finds no treatment effect (p = 0.44) with parallel pre-trends, matching leaves the gap unexplained by size or profitability, and adopters are about ten times larger—indicating selection, not treatment. Tool type does not matter, while non-adopters' efficiency falls monotonically with the carbon price (cumulative −23.1%/−19.3%). We reposition ICP as a risk-management and signaling instrument rather than a performance scorecard.

Ajala, A. A., Adeoye, O. L., Salami, O. M., & Jimoh, A. Y. (2025). An examination of daily CO2 emissions prediction through a comparative analysis of machine learning, deep learning, and statistical models. Environmental Science and Pollution Research, 32(5), 2510–2535.
Aldy, J. E., & Gianfrate, G. (2019). Future-proof your climate strategy. Harvard Business Review, 97(3), 86–97.
Ambec, S., Cohen, M. A., Elgie, S., & Lanoie, P. (2013). The Porter hypothesis at 20: Can environmental regulation enhance innovation and competitiveness? Review of Environmental Economics and Policy, 7(1), 2–22.
Athey, S., & Imbens, G. W. (2019). Machine learning methods that economists should know about. Annual Review of Economics, 11, 685–725.
Banker, R. D., & Natarajan, R. (2008). Evaluating contextual variables affecting productivity using data envelopment analysis. Operations Research, 56(1), 48–58.
Ben-Amar, W., Gomes, M., Khursheed, H., & Marsat, S. (2022). Climate change exposure and internal carbon pricing adoption. Business Strategy and the Environment, 31(7), 2854–2870.
Bento, N., & Gianfrate, G. (2020). Determinants of internal carbon pricing. Energy Policy, 143, 111499.
Bolton, P., & Kacperczyk, M. (2021). Do investors care about carbon risk? Journal of Financial Economics, 142(2), 517–549.
Çakır, S. (2024). Best output prediction in OECD railways using DEA in conjunction with machine learning algorithms. Annals of Operations Research, 335(1), 59–77.
Callaway, B., & Sant’Anna, P. H. C. (2021). Difference-in-differences with multiple time periods. Journal of Econometrics, 225(2), 200–230.
CDP. (2021). Putting a price on carbon: The state of internal carbon pricing by corporates globally. CDP Worldwide.
Charnes, A., Cooper, W. W., & Rhodes, E. (1978). Measuring the efficiency of decision making units. European Journal of Operational Research, 2(6), 429–444.
Chen, L., & Wang, S. (2024). Directional distance functions data envelopment analysis method with endogenous direction for target setting. Journal of the Operational Research Society, 75(9), 1699–1710.
Cho, K., van Merriënboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014). Learning phrase representations using RNN encoder–decoder for statistical machine translation. arXiv:1406.1078.
Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014). Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555.
Chung, Y. H., Färe, R., & Grosskopf, S. (1997). Productivity and undesirable outputs: A directional distance function approach. Journal of Environmental Management, 51(3), 229–240.
Clarkson, P. M., Li, Y., Richardson, G. D., & Vasvari, F. P. (2008). Revisiting the relation between environmental performance and environmental disclosure: An empirical analysis. Accounting, Organizations and Society, 33(4–5), 303–327.
Cooper, W. W., Seiford, L. M., & Zhu, J. (2011). Handbook on data envelopment analysis (2nd ed.). Springer.
Cubric, M. (2020). Drivers, barriers and social considerations for AI adoption in business and management: A tertiary study. Technology in Society, 62, 101257.
Daraio, C., & Simar, L. (2005). Introducing environmental variables in nonparametric frontier models: A probabilistic approach. Journal of Productivity Analysis, 24(1), 93–121.
de Chaisemartin, C., & D’Haultfœuille, X. (2020). Two-way fixed effects estimators with heterogeneous treatment effects. American Economic Review, 110(9), 2964–2996.
Dixit, A. K., & Pindyck, R. S. (1994). Investment under uncertainty. Princeton University Press.
Färe, R., Grosskopf, S., & Lovell, C. A. K. (1994). Production frontiers. Cambridge University Press.
Färe, R., Grosskopf, S., Lovell, C. A. K., & Pasurka, C. (1989). Multilateral productivity comparisons when some outputs are undesirable: A nonparametric approach. The Review of Economics and Statistics, 71(1), 90–98.
Färe, R., Grosskopf, S., Lovell, C. A. K., & Yaisawarng, S. (1993). Derivation of shadow prices for undesirable outputs: A distance function approach. The Review of Economics and Statistics, 75(2), 374–380.
Freeman, R. E. (1984). Strategic management: A stakeholder approach. Pitman.
Goodman-Bacon, A. (2021). Difference-in-differences with variation in treatment timing. Journal of Econometrics, 225(2), 254–277.
Gorbach, O. G., Kost, C., & Pickett, C. (2022). Review of internal carbon pricing and the development of a decision process for the identification of promising internal pricing methods for an organisation. Renewable and Sustainable Energy Reviews, 154, 111745.
Green, J. F. (2021). Does carbon pricing reduce emissions? A review of ex-post analyses. Environmental Research Letters, 16(4), 043004.
Gu, S., Kelly, B., & Xiu, D. (2020). Empirical asset pricing via machine learning. The Review of Financial Studies, 33(5), 2223–2273.
Harpankar, K. (2019). Internal carbon pricing: Rationale, promise and limitations. Carbon Management, 10(2), 219–225.
Hewamalage, H., Bergmeir, C., & Bandara, K. (2021). Recurrent neural networks for time series forecasting: Current status and future directions. International Journal of Forecasting, 37(1), 388–427.
Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780.
Hoff, A. (2007). Second stage DEA: Comparison of approaches for modelling the DEA score. European Journal of Operational Research, 181(1), 425–435.
Hsu, P.-H., Li, K., & Tsou, C.-Y. (2023). The pollution premium. The Journal of Finance, 78(3), 1343–1392.
Ilhan, E., Sautner, Z., & Vilkov, G. (2021). Carbon tail risk. The Review of Financial Studies, 34(3), 1540–1571.
In, S. Y., Park, K. Y., & Monk, A. H. B. (2019). Is ‘being green’ rewarded in the market? An empirical investigation of decarbonization and stock returns (Stanford Global Project Center Working Paper). SSRN. https://ssrn.com/abstract=3020304
Jin, Y., Sharifi, A., Li, Z., Chen, S., Zeng, S., & Zhao, S. (2024). Carbon emission prediction models: A review. Science of the Total Environment, 927, 172319.
Kao, C. (2014). Network data envelopment analysis: A review. European Journal of Operational Research, 239(1), 1–16.
Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T.-Y. (2017). LightGBM: A highly efficient gradient boosting decision tree. Advances in Neural Information Processing Systems, 30, 3146–3154.
Krueger, P., Sautner, Z., & Starks, L. T. (2020). The importance of climate risks for institutional investors. The Review of Financial Studies, 33(3), 1067–1111.
Lilliestam, J., Patt, A., & Bersalli, G. (2021). The effect of carbon pricing on technological change for full energy decarbonization: A review of empirical ex-post evidence. WIREs Climate Change, 12(1), e681.
Liu, H., & Shen, L. (2020). Forecasting carbon price using empirical wavelet transform and gated recurrent unit neural network. Carbon Management, 11(1), 25–37.
Ma, J., & Kuo, J. (2021). Environmental self-regulation for sustainable development: Can internal carbon pricing enhance financial performance? Business Strategy and the Environment, 30(8), 3517–3527.
Makridakis, S., Spiliotis, E., & Assimakopoulos, V. (2018). Statistical and machine learning forecasting methods: Concerns and ways forward. PLOS ONE, 13(3), e0194889.
Matsumura, E. M., Prakash, R., & Vera-Muñoz, S. C. (2014). Firm-value effects of carbon emissions and carbon disclosures. The Accounting Review, 89(2), 695–724.
McDonald, J. (2009). Using least squares and tobit in second stage DEA efficiency analyses. European Journal of Operational Research, 197(2), 792–798.
Mergoni, A., Emrouznejad, A., & De Witte, K. (2024). Fifty years of data envelopment analysis. European Journal of Operational Research, 313(1), 1–25.
Microsoft. (2022, March 24). How Microsoft is using an internal carbon fee to reach its carbon negative goal. Microsoft Industry Blog. https://www.microsoft.com/en-us/industry/blog/sustainability/2022/03/24/how-microsoft-is-using-an-internal-carbon-fee-to-reach-its-carbon-negative-goal/
Modhej, D., Sanei, M., Shoja, N., & Hosseinzadeh Lotfi, F. (2017). Integrating inverse data envelopment analysis and neural network to preserve relative efficiency values. Journal of Intelligent & Fuzzy Systems, 32(6), 4047–4058.
Mullainathan, S., & Spiess, J. (2017). Machine learning: An applied econometric approach. Journal of Economic Perspectives, 31(2), 87–106.
Nguyen, Q., Diaz-Rainey, I., & Kuruppuarachchi, D. (2021). Predicting corporate carbon footprints for climate finance risk analyses: A machine learning approach. Energy Economics, 95, 105129.
Nishant, R., Kennedy, M., & Corbett, J. (2020). Artificial intelligence for sustainability: Challenges, opportunities, and a research agenda. International Journal of Information Management, 53, 102104.
Pástor, Ľ., Stambaugh, R. F., & Taylor, L. A. (2021). Sustainable investing in equilibrium. Journal of Financial Economics, 142(2), 550–571.
Pástor, Ľ., Stambaugh, R. F., & Taylor, L. A. (2022). Dissecting green returns. Journal of Financial Economics, 146(2), 403–424.
Porter, M. E., & van der Linde, C. (1995). Toward a new conception of the environment-competitiveness relationship. Journal of Economic Perspectives, 9(4), 97–118.
Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70(1), 41–55.
Seiford, L. M., & Zhu, J. (2002). Modeling undesirable factors in efficiency evaluation. European Journal of Operational Research, 142(1), 16–20.
Sharp, J. A., Meng, W., & Liu, W. (2007). A modified slacks-based measure model for data envelopment analysis with ‘natural’ negative outputs and inputs. Journal of the Operational Research Society, 58(12), 1672–1677.
Simar, L., & Wilson, P. W. (1998). Sensitivity analysis of efficiency scores: How to bootstrap in nonparametric frontier models. Management Science, 44(1), 49–61.
Simar, L., & Wilson, P. W. (2007). Estimation and inference in two-stage, semi-parametric models of production processes. Journal of Econometrics, 136(1), 31–64.
Simar, L., & Wilson, P. W. (2011). Two-stage DEA: Caveat emptor. Journal of Productivity Analysis, 36(2), 205–218.
Smith, B. (2020, January 16). Microsoft will be carbon negative by 2030. The Official Microsoft Blog. https://blogs.microsoft.com/blog/2020/01/16/microsoft-will-be-carbon-negative-by-2030/
Smith, B. (2021, January 28). One year later: The path to carbon negative—a progress report on our climate ‘moonshot’. The Official Microsoft Blog. https://blogs.microsoft.com/blog/2021/01/28/one-year-later-the-path-to-carbon-negative-a-progress-report-on-our-climate-moonshot/
Stiglitz, J. E., & Stern, N. (2017). Report of the High-Level Commission on Carbon Prices. World Bank.
Stock, J. H., & Watson, M. W. (2009). Forecasting in dynamic factor models subject to structural instability. In J. Castle & N. Shephard (Eds.), The methodology and practice of econometrics: A festschrift in honour of David F. Hendry (pp. 173–205). Oxford University Press.
Suchman, M. C. (1995). Managing legitimacy: Strategic and institutional approaches. Academy of Management Review, 20(3), 571–610.
Sueyoshi, T., & Goto, M. (2012). DEA environmental assessment of coal fired power plants: Methodological comparison between radial and non-radial models. Energy Economics, 34(6), 1854–1863.
Sueyoshi, T., & Goto, M. (2018). Environmental assessment on energy and sustainability by data envelopment analysis. Wiley.
Sueyoshi, T., Yuan, Y., & Goto, M. (2017). A literature study for DEA applied to energy and environment. Energy Economics, 62, 104–124.
Sun, L., & Abraham, S. (2021). Estimating dynamic treatment effects in event studies with heterogeneous treatment effects. Journal of Econometrics, 225(2), 175–199.
Tone, K. (2001). A slacks-based measure of efficiency in data envelopment analysis. European Journal of Operational Research, 130(3), 498–509.
Tone, K. (2004). Dealing with undesirable outputs in DEA: A slacks-based measure (SBM) approach. Presentation at NAPW III, Toronto, 44–45.
Tone, K., & Tsutsui, M. (2010). Dynamic DEA: A slacks-based measure approach. Omega, 38(3–4), 145–156.
Tone, K., & Tsutsui, M. (2014). Dynamic DEA with network structure: A slacks-based measure approach. Omega, 42(1), 124–131.
Trinks, A., Mulder, M., & Scholtens, B. (2022). External carbon costs and internal carbon pricing. Renewable and Sustainable Energy Reviews, 168, 112780.
Tsolas, I. E., Charles, V., & Gherman, T. (2020). Supporting better practice benchmarking: A DEA-ANN approach to bank branch performance assessment. Expert Systems with Applications, 160, 113599.
Varian, H. R. (2014). Big data: New tricks for econometrics. Journal of Economic Perspectives, 28(2), 3–28.
Verrecchia, R. E. (1983). Discretionary disclosure. Journal of Accounting and Economics, 5, 179–194.
Wang, H., Lei, Z., Zhang, X., Zhou, B., & Peng, J. (2019). A review of deep learning for renewable energy forecasting. Energy Conversion and Management, 198, 111799.
Wang, Q., Zhou, P., & Zhou, D. (2012). Efficiency measurement with carbon dioxide emissions: The case of OECD countries. Applied Energy, 92, 161–170.
World Bank. (2023). State and trends of carbon pricing 2023. World Bank.
Zhang, N., & Choi, Y. (2014). A note on the evolution of directional distance function and its development in energy and environmental studies 1997–2013. Renewable and Sustainable Energy Reviews, 33, 50–59.
Zhou, P., Ang, B. W., & Poh, K. L. (2008). A survey of data envelopment analysis in energy and environmental studies. European Journal of Operational Research, 189(1), 1–18.
Zhu, N., Zhu, C., & Emrouznejad, A. (2021). A combined machine learning algorithms and DEA method for measuring and predicting the efficiency of Chinese manufacturing listed companies. Journal of Management Science and Engineering, 6(4), 435–448.