Every industrial enterprise keeps three separate accounts of the same goods. The warehouse floor holds a physical record of what actually moved. The enterprise system holds a transactional record of what was documented and priced. The tax function holds a fiscal record of what was declared to the authorities. A new conceptual analysis published in Frontiers in Big Data argues that indirect taxation works, or fails, precisely to the extent that these three records agree with one another, and that this agreement, rather than any single ledger, should be treated as the true object of tax accountability.
The article, authored by researchers affiliated with universities in Karshi, Uzbekistan, and elsewhere, is a conceptual analysis rather than an empirical study. Its central observation is that the standard evasion schemes in manufacturing, namely unrecorded output, invented purchasing, disguised disposal and elastic valuation, all operate by driving the three records apart while leaving each one internally plausible. A factory can understate production in its fiscal declarations while its physical stock movements tell a different story; it can invent purchases that exist only on paper; it can route goods to informal disposal without documentation; and it can manipulate declared values within plausible ranges. Each scheme corrupts the relationship between records without necessarily corrupting any single record.
What has changed, the authors argue, is the technological substrate on which these records now sit. Serialised marking of goods, sensor networks, enterprise event logs and clearance-model electronic invoicing, in which an invoice must be validated by a tax authority before it takes effect, mean that the three accounts now exist at population scale and in machine-readable form. Discrepancies that once required laborious manual audit sampling to uncover can in principle be tested continuously and comprehensively across entire economies.
The paper draws on tax compliance economics, big data research in auditing, continuous assurance literature and organisational capability theory to reconstruct accountability as a property of agreement among records. But it is careful to state the conditions under which agreement is actually evidence of anything. A reconciliation between two records carries weight only where its two sides descend from causally separate processes. If both sides of a comparison are generated inside the same integrated enterprise transaction-processing layer, agreement between them may simply reflect a shared software pipeline rather than independent confirmation of physical reality. Integrated systems, the authors conclude, violate this separability condition by default, which is why records must contain measurement anchors generated outside that layer, such as sensor readings or serialised scans captured at the point of physical movement.
Having established that condition, the article formalises agreement as a set of reconciliation residuals, the differences between paired records, and derives a detection condition from stated assumptions. The core result is stark: a scheme survives only while its divergence stays inside the noise floor of the tightest independent check that sees it. From this the authors derive that the sustainable tax gap is the sum of those noise floors rather than a function of the number of controls in place. Adding more controls that all observe the same correlated data adds little; tightening the tolerance of an independent check shrinks the gap directly.
A sensitivity analysis examines how this bound moves when the assumptions are relaxed. The model assumes separability of record-generating processes, exogenous tolerances and zero detection risk; the analysis shows how the sustainable gap expands as each of these assumptions weakens, for example when tolerances become negotiable in practice or when detection carries probabilistic rather than certain consequences.
The framework is then applied to four leakage channels traced through the inventory cycle: unrecorded output, invented purchasing, disguised disposal and elastic valuation. Each channel is matched to the specific reconciliation that exposes it, and the analysis is condensed into four propositions, each accompanied by refuting observations and candidate measures. This structure is intended to make the framework testable rather than merely descriptive.
Uzbekistan serves as the instantiating setting for the argument. The country has deployed clearance invoicing and serialised labelling, and from January 2026 is introducing automated invoice risk scoring. The authors argue that these state-side digital capabilities have outrun the enterprise-side capability of many firms to maintain records that can withstand population-scale reconciliation. The paper also draws on documented administrative reconciliation practice in Uzbekistan between 2021 and 2023, which supplies what the authors call a live episode and its governance failure mode, illustrating how reconciliation processes can exist on paper yet fail in operation.
The broader implication is a shift in how tax compliance should be engineered. Rather than auditing documents one at a time, authorities and firms alike can focus on the independence and precision of the checks that connect physical, transactional and fiscal records. Where those checks are causally independent and their noise floors are narrow, the room for internally plausible evasion shrinks to the residual noise itself. Where they are not, additional digital controls may create the appearance of assurance without constraining the gap. The article, received in July 2026 and accepted in October 2026, positions record agreement as a measurable, designable property of fiscal systems in the era of population-scale digital data.
Subject of Research: A conceptual framework treating indirect tax accountability as agreement among physical, transactional and fiscal records, tested against digital invoicing and inventory technologies in Uzbekistan.
Article Title: Indirect tax accountability as record agreement: inventory accounting and auditing under digital technologies and big data
Article References: Alimkhanova, N., Djalilov, R. H., Egamberdieva, S. R., Ismatullaev, J. A., Kholmuradov, O. N., Oramova, N., Khushmuradov, O., Anvarova, L., & Raimberdiev, A. A. (2026). Indirect tax accountability as record agreement: inventory accounting and auditing under digital technologies and big data. Frontiers in Big Data, 9, Article 1953628. https://doi.org/10.3389/fdata.2026.1953628
Image Credits: AI Generated
DOI: 10.3389/fdata.2026.1953628
Keywords: indirect taxation, value added tax, tax compliance, inventory accounting, big data, continuous auditing, electronic invoicing, serialized marking, reconciliation residuals, Uzbekistan, tax gap, digital records
News Source: Denise Maddox. (October 28, 2026). Three Records, One Tax: When Digital Ledgers Agree, Evasion Runs Out of Room. Scienmag.
