Beyond Black-Box Budgeting: An Explainable Deep Learning Framework for Fiscal Impact Prediction and Resource Allocation in Multi-Level Government Systems
DOI:
https://doi.org/10.54938/ijemdss.2026.05.4.823Keywords:
Explainable AI, Fiscal Forecasting, Multi-Level Governance, Deep Learning, Resource Allocation, Block-Additive Models, Hierarchical Time Series, Public Financial Management, SHAP, Performance-Based BudgetingAbstract
The growing complexity of multi-level governance systems, which entails complex intergovernmental fiscal flows, uneven socioeconomic situations, and the rising demands of fiscal transparency has revealed the shortcomings of the traditional budgetary systems. This paper presents a theoretical framework and demonstrates the validity of an explainable deep learning framework that goes beyond traditional black-box methods, by combining hybrid neural networks with theoretically-based interpretability systems. The proposed framework is a Multi-Level Hierarchical Attention Network (ML-HAN) which integrates 1D-CNNs to extract features, bi-directional LSTMs to capture temporal dependencies, and a new hierarchical attention that allows fiscal impact attribution on various levels of government. It makes use of Block-Additive Structures (BAMs) that provide economically meaningful constraints, post-hoc explanations based on SHAP, and axiom-constrained resource allocation optimization. A panel analysis over 15 years of 3,142 counties in the United States and 50 states demonstrates that our model has a better predictive performance (RMSE = 3.87, MAPE = 4.2) than traditional econometric models (ARDL: RMSE = 7.91; NARDL: RMSE = 6.84) and offers a finer, policy-actionable detail of fiscal determinants. The resource allocation module of the framework is proven by counterfactual simulation to enhance allocating efficiency by 23.7 percent compared to the historical trend. We find that explainable AI can help overcome the disconnect between predictive capacity and policy responsibility in governmental financial management, and provide an avenue toward intelligent, transparent, and equitable fiscal management.
Downloads
References
Adadi, A., & Berrada, M. (2018). Peeking inside the black-box: A survey on explainable artificial intelligence (XAI). IEEE Access, 6, 52138–52160.
Andrews, M. (2020). Budgeting and performance: A practical guide for public sector organizations. Cambridge University Press.
Atkinson, A. B. (1970). On the measurement of inequality. Journal of Economic Theory, 2(3), 244–263.
Aziz, H., & Shah, N. (2021). Participatory budgeting: Models and approaches. In Pathways Between Social Science and Computational Social Science (pp. 215–236). Springer.
Azmi, S. K., Vethachalam, S., & Karamchand, G. (2025). Predictive analytics for national budgeting and expenditure: Leveraging AI/ML on the PFMS 2.0 data ecosystem. Journal of Information Systems Engineering and Management, 10(2), 2025–2036.
Barredo Arrieta, A., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., ... & Herrera, F. (2020). Explainable artificial intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information Fusion, 58, 82–115.
Baskaran, T. (2010). On the link between fiscal decentralization and public debt in OECD countries. Public Choice, 145(3), 351–378.
Bhuiyan, N. H., Islam, D., Hossain, A., & Islam, M. R. (2026). An intelligent data-driven framework for transparent government support distribution in Bangladesh using ML and XAI. In Proceedings of the International Conference on Intelligent Data Analysis and Applications (IDAA 2025) (pp. 1362–1377). Atlantis Press.
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.
Bucci, D. (2025). Fiscal impulse responses in a high-dimensional setting. CFE-CMStatistics 2025 Conference Proceedings.
Buckmann, M., & Potjagailo, G. (2025). Infusing economically motivated structure into machine learning methods. Bank of England Staff Working Paper No. 1,144.
Caruana, R., Lou, Y., Gehrke, J., Koch, P., Sturm, M., & Elhadad, N. (2015). Intelligible models for healthcare: Predicting pneumonia risk and hospital 30-day readmission. In Proceedings of the 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1721–1730).
Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785–794).
Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C., Newey, W., & Robins, J. (2018). Double/debiased machine learning for treatment and structural parameters. The Econometrics Journal, 21(1), C1–C68.
Clements, M. P. (2023). Machine learning and economic forecasting. Journal of Economic Surveys, 37(3), 1–35.
Diebold, F. X., & Mariano, R. S. (1995). Comparing predictive accuracy. Journal of Business & Economic Statistics, 13(3), 253–263.
Drucker, H., Burges, C. J. C., Kaufman, L., Smola, A., & Vapnik, V. (1997). Support vector regression machines. In Advances in Neural Information Processing Systems (Vol. 9, pp. 155–161).
Eubanks, V. (2018). Automating inequality: How high-tech tools profile, police, and punish the poor. St. Martin's Press.
Gal, Y., & Ghahramani, Z. (2016). Dropout as a Bayesian approximation: Representing model uncertainty in deep learning. In Proceedings of the 33rd International Conference on Machine Learning (pp. 1050–1059).
Gore, A. K. (2018). Does local government fiscal distress cause municipal bankruptcy? Public Administration Review, 78(1), 17–28.
Guidotti, R., Monreale, A., Ruggieri, S., Turini, F., Giannotti, F., & Pedreschi, D. (2018). A survey of methods for explaining black box models. ACM Computing Surveys, 51(5), 1–42.
Hamilton, J. D. (1994). Time series analysis. Princeton University Press.
Hansen, P. R., Lunde, A., & Nason, J. M. (2011). The model confidence set. Econometrica, 79(2), 453–497.
Hashemi, M., Darejeh, A., & Cruz, F. (2024). A user-centric exploration of axiomatic explainable AI in participatory budgeting. In Proceedings of the 2024 ACM Conference on AI and Public Administration (pp. 1–10). ACM.
Hines, J. R., & Thaler, R. H. (1995). Anomalies: The flypaper effect. Journal of Economic Perspectives, 9(4), 217–226.
Hyndman, R. J., Ahmed, R. A., Athanasopoulos, G., & Shang, H. L. (2011). Optimal combination forecasts for hierarchical time series. Computational Statistics & Data Analysis, 55(9), 2579–2589.
Hyndman, R. J., & Athanasopoulos, G. (2021). Forecasting: Principles and practice (3rd ed.). OTexts.
Jawad, S. H., Hmood, M. Y., & Matrood, Z. Y. (2026). Predicting public budget surplus and deficit using a hybrid 1D-CNN–LSTM model. Statistics, Optimization & Information Computing, 16(2), 1143–1159.
Joyce, P. G. (2003). Linking performance and budgeting: Opportunities in the federal budget process. IBM Center for the Business of Government.
Karamchand, G. (2025). Artificial intelligence as a driving force for performance-based budgeting with a real-time reporting approach in the public sector. In Proceedings of the International Conference on Public Financial Management.
Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. In International Conference on Learning Representations (ICLR).
Koop, G., & Korobilis, D. (2010). Bayesian multivariate time series methods for empirical macroeconomics. Foundations and Trends in Econometrics, 3(4), 267–358.
Künzel, S. R., Sekhon, J. S., Bickel, P. J., & Yu, B. (2019). Metalearners for estimating heterogeneous treatment effects using machine learning. Proceedings of the National Academy of Sciences, 116(10), 4156–4165.
Liu, R., Li, H., Yoon, K., & Vasarhelyi, M. A. (2025). Using machine learning algorithms to improve fiscal distress prediction models: The case of U.S. local governments. Journal of Information Systems, 39(3), 131–155.
Loshchilov, I., & Hutter, F. (2019). Decoupled weight decay regularization. In International Conference on Learning Representations (ICLR).
Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems (Vol. 30, pp. 4765–4774).
Lundberg, S. M., & Lee, S.-I. (2018). Consistent feature attribution for tree ensembles. In Proceedings of the 34th International Conference on Machine Learning (pp. 3146–3155).
Medeiros, M. C., Vasconcelos, G. F. R., & Veiga, A. (2021). Forecasting inflation in a data-rich environment: The benefits of machine learning methods. Journal of Business & Economic Statistics, 39(1), 98–119.
Melkers, J., & Willoughby, K. (2001). Budgeters' views of state performance-budgeting systems: Distinctions across branches. Public Administration Review, 61(1), 54–64.
Mirrlees, J. A. (1971). An exploration in the theory of optimum income taxation. Review of Economic Studies, 38(2), 175–208.
Moynihan, D. P., & Pandey, S. K. (2010). The big question for performance management: Why do managers use performance information? Journal of Public Administration Research and Theory, 20(4), 849–866.
Musgrave, R. A. (1959). The theory of public finance. McGraw-Hill.
Neto, J. I. C., Lima Neto, F. B., & de Oliveira, J. F. L. (2022). Use of machine learning and multilevel analysis in hierarchical approaches of public expenditure forecasting. In 2022 IEEE International Conference on Systems, Man, and Cybernetics (SMC) (pp. 1155–1160).
Oates, W. E. (1972). Fiscal federalism. Harcourt Brace Jovanovich.
Oates, W. E. (1999). An essay on fiscal federalism. Journal of Economic Literature, 37(3), 1120–1149.
O'Neil, C. (2016). Weapons of math destruction: How big data increases inequality and threatens democracy. Crown Publishers.
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., ... & Chintala, S. (2019). PyTorch: An imperative style, high-performance deep learning library. In Advances in Neural Information Processing Systems (Vol. 32, pp. 8026–8037).
Pattison, N. (2017). Financial condition of U.S. local governments: A review of the literature. Government Finance Review.
Pesaran, M. H., & Shin, Y. (1999). An autoregressive distributed-lag modelling approach to cointegration analysis. In Econometrics and Economic Theory in the 20th Century (pp. 371–413). Cambridge University Press.
Rawls, J. (1971). A theory of justice. Harvard University Press.
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). "Why should I trust you?": Explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1135–1144).
Rodden, J. (2004). Comparative federalism and decentralization: On meaning and measurement. Comparative Politics, 36(4), 481–500.
Saez, E., & Stantcheva, S. (2016). Generalized social marginal welfare weights for optimal tax theory. American Economic Review, 106(1), 24–45.
Samek, W., Wiegand, T., & Müller, K.-R. (2017). Explainable artificial intelligence: Understanding, visualizing and interpreting deep learning models. IEEE Signal Processing Magazine, 34(6), 42–48.
Schick, A. (2014). Budgeting and fiscal management in the twenty-first century. In The Oxford Handbook of Public Accountability (pp. 416–435). Oxford University Press.
Schuknecht, L. (2005). Political business cycles in EU countries. European Journal of Political Economy, 21(1), 91–110.
Sharma, J., Gupta, P., & Khanna, N. (2025). Moving beyond black boxes with interpretable anomaly detection in government finances. In 2025 IEEE 5th International Conference on ICT in Business Industry & Government (ICTBIG) (pp. 1–7).
Shin, Y., Yu, B., & Greenwood-Nimmo, M. (2014). Modelling asymmetric cointegration and dynamic multipliers in a nonlinear ARDL framework. In Festschrift in Honor of Peter Schmidt (pp. 281–314). Springer.
Stock, J. H., & Watson, M. W. (2001). Vector autoregressions. Journal of Economic Perspectives, 15(4), 101–115.
Tarasiuk, M. (2025). Intellectualization of public sector financial resource management: Analytical tools and efficiency models. Financial and Credit Systems: Prospects for Development, 4(19), 265–283.
Tukey, J. W. (1977). Exploratory data analysis. Addison-Wesley.
van Buuren, S. (2018). Flexible imputation of missing data (2nd ed.). CRC Press.
Vandervieren, E., & Hubert, M. (2008). Boxplot-based outlier detection for skewed distributions. Computational Statistics & Data Analysis, 52(12), 5186–5201.
Venkatasubbu, P., Patil, R., & Jaincy, M. (2025). A novel AI-driven framework for adaptive investment decision-making using India's fiscal budget signals. In 2025 International Conference on Sensors and Related Networks (SENNET) (pp. 1–6).
Wang, X., Li, J., & Liu, Y. (2023). Deep learning for time series forecasting: A survey. IEEE Transactions on Neural Networks and Learning Systems, 34(5), 2012–2032.
Wildavsky, A., & Caiden, N. (2004). The new politics of the budgetary process (5th ed.). Longman.
Zinkevich, M. (2003). Online convex programming and generalized infinitesimal gradient ascent. In Proceedings of the 20th International Conference on Machine Learning (pp. 928–936).
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Muhammad Usman Malik, Um-e-Laila, Ahsan Ullah , Muhammad Aqeel, Muhammad Wasim
This work is licensed under a Creative Commons Attribution 4.0 International License.
Under the Creative Common Attribution (CC-BY 4.0) license, authors retain copyright and grant the journal right of first publication.
