New Technologies in International Law / Tymofeyeva, Crhák et al.

A more extensive study has been conducted in Tajikistan. Using data from small and medium-sized firms in Dushanbe, according to the study, e-filing dramatically saved the time spent on tax-related operations by 40%, effectively tripling taxes paid by enterprises more likely to evade while decreasing payments by firms less likely to avoid. E-filing also resulted in fewer bribes because extortion chances were eliminated. 374 A study conducted in Peru investigated the impact of electronic invoicing reforms on tax compliance and technology adoption. It found that firms are more likely to adopt e-invoicing voluntarily if their trading partners are mandated to do so, demonstrating positive spillover effects in technology adoption. 375 2.3 Big Data and Analytics Big data involves processing large volumes of diverse information quickly and cost effectively for improved knowledge, decision-making, and automation. It encompasses three key dimensions, volume, speed, and variety. Large-scale data from multiple sources is combined with sophisticated algorithms and uses methods like data mining, machine learning, and neural networks. 376 A lot of policy discussion surrounds investments in detection capabilities, such as third-party reporting, the use of electronic databases, and technology-based tools for comprehending and tracking the tax base. 377 Some countries use datamining or machine learning systems based on artificial intelligence to support their tax audits, to identify fraud risks in an automated way, or to carry out research, investigation, programming, control and recovery operations for tax violations. In Brazil, the use of artificial intelligence and big data has allowed, even during the pandemic, to increase the Brazilian budget by 10%. 378 By evaluating massive datasets to spot irregularities, like differences in sales reporting for income tax and VAT, technology improves tax monitoring. 379 In Chile, the Servicio de Impuestos Internos (SII) employs AI tools to predict which taxpayers are most likely to underreport income, thereby optimizing their audit selection process. 380 Another recent development in Argentina demonstrates how tax authorities must adapt to taxpayer behavior. In 2022, Argentina’s tax administration (AFIP) implemented the Comprehensive System for Monitoring Payments Abroad for Services (SIMPES), which has been effective in discovering 374 Okunogbe O and Pouliquen V, ‘Technology, Taxation, and Corruption: Evidence from the Introduction of Electronic Tax Filing.’ (2022) 14(1) American Economic Journal: Economic Policy 341. 375 Bellon M, ‘Technology and Tax Compliance Spillovers: Evidence from a VAT E-Invoicing Reform in Peru’ (2023) Journal of Economic Behavior & Organization 212, p. 756. 376 Faúndez A, Mellado-Silva R and Aldunate-Lizana E, ‘Use of Artificial Intelligence by Tax Administrations: An Analysis Regarding Taxpayers’ Rights in Latin American Countries’ (2020) 38 Computer Law & Security Review 105441, p. 3. 377 Okunogbe O, ‘Becoming Legible to the State: The Role of Identification and Collection Capacity in Taxation (English)’ ( World Bank Group , 2021) , p. 19. 378 ILA, Sadowsky, p. 70. 379 ‘The Promise and Limitations of Information Technology for Tax Mobilization’ (2023), p. 317. 380 ‘Use of Artificial Intelligence by Tax Administrations: An Analysis Regarding Taxpayers’ Rights in Latin American Countries’, pp. 3–4.

96