A data control framework for SAF-T reporting: A process-based approach.

• Author: Auksztol, Jerzy

INTRODUCTION

The Standard Audit File for Tax (SAF-T) is an open standard format based on eXtensible Markup Language (XML), which enables the storing and transferring of data for tax audit purposes. It was first published by the Organization for Economic Co-operation and Development (OECD) in 2005, with major amendments and extensions in 2010 (OECD, 2017b, p. 65). Revenue bodies in many countries have adopted this standard with great success, inter alia in Austria, France, Luxembourg, Lithuania, Norway, Poland, and Portugal. The relatively simple idea of sending financial records directly to the tax authorities has changed significantly the approach to audits on tax compliance. Its characteristics can be grouped in a four-fold way: (1) the reduction of burden by automation, (2) moving from collecting aggregates to individual records, (3) tax compliance process redesign, (4) increasing the taxpayer's responsibility for evaders' misconduct.

1) The reduction of burden by automation. It reflects the postulate to reduce tax compliance cost to a minimum (Braithwaite, 2013a, p. 1). SAF-T can optimize efficiently the mutual communication between taxpayers and revenue bodies. On the taxpayers' side, it is achieved by changing the features of the enterprise information system (EIS) (4), where the functionality of exporting financial data to SAF-T has been built in by EIS developers. Therefore, this software can generate the required records and file them with the tax authorities via the Internet. The taxpayer's manual data feed is not needed anymore. On the opposite end, the tax administration automatically receives financial records from taxpayers and loads them into the administrative databases for further control, analysis and audit, eliminating manual or semi-manual, time-consuming work.

2) Moving from collecting aggregates to individual records. SAF-T extends significantly the information scope available to revenue bodies without unnecessary delay. It helps build administrative registers consisting of individual records taken from different databases kept by taxpayers. For example, SAF-T for VAT collects data for each transaction carried out by VAT payers, which can be matched with corresponding records from their contractors. In consequence, the revenue body receives powerful tools to build a pool of cases earmarked for more precise and accurate fiscal auditing.

3) Tax compliance process redesign. Built by revenue bodies over the years, the system of tax audit based on collecting aggregates has changed significantly, mainly due to the possibility of direct access to the individual records of the whole population of taxpayers. Choosing individuals for tax audit has become much more precise since the introduction of SAFT. Big Data technologies (Vasarhelyi, Kogan, & Tuttle, 2015) and Business Intelligence analytics (Wu, Ou, Lin, Chand, & Yen, 2012; OECD, 2017a, p. 38) which have been adopted by tax administration are also helpful in this field. This also opens the perspective for mitigating the negative effects of tax audit where a subjective interpretation may cause tension between an honest taxpayer and an inflexible tax inspector (Kirchler, Hoelzl, & Wahl, 2008, p. 214-215).

4) Increasing the taxpayer's responsibility for evaders' misconduct. It mainly results from the very nature of value added tax (VAT), where (Ebrill, Keen, & Bodin, 2001, p. 2) the tax is charged and collected throughout the production process, with provision for tax payable to be reduced by the tax paid in respect of purchases. It means that the tax paid by the final consumer is collected at each stage of the production and distribution chain. Taxpayers can reduce their tax by the amount paid earlier. This encourages sophisticated VAT frauds, which can involve honest taxpayers. It happens through the reduction of tax payable by the amount of tax shown in the documentation of the purchasing process, but in reality not paid by tax evaders and fraudsters. One of the methods of VAT fraud is called missing trader or carousel, discussed extensively in specialist literature (Ainsworth, 2006; Keen & Smith, 2006). Another well-known mechanism is connected with the carbon allowance (Frunza, Guegan, & Tchiebaut, 2010, p. 4), which can be traded on an electronic exchange market, thus increasing the fraud speed and volume. Tax authorities, in response to these kinds of frauds, have introduced changes to legislation that impose the tax liability on all participants involved in a transaction, even when some of them are unaware of the tax evasion. Collecting individual transaction records through SAF-T increases the compliance due to more accurate detection of all participants in fraudulent transactions.

This paper has two main objectives. The first one is to examine the accountants' attitude to changes caused by the introduction of the breakthrough SAF-T reporting. Accountants bear overall responsibility for tax compliance and build their own internal, accounting-focused procedures for fulfilling this task. Now, the changed circumstances have triggered a need to spread the responsibility for tax compliance to key employees performing processes involving vendors and purchasers. The accountants' attitude in this matter can be a motivation for research in the management field.

The second aim of this paper is to explore a Data Control Framework (DCF) for SAF-T, which helps entrepreneurs to build a resilient organization which responds in an agile way to the new circumstances arising from the introduction of SAF-T. Thus, we raise a discussion about the impact of SAF-T on decision processes executed at all levels of the economy: micro, mezzo, and macro.

The main body of our study is divided into seven parts. Following the introduction, we identify the problem to be solved and set up the overall goal of DCF in the context of changing the relationship between entrepreneurs and tax authorities. In the third section, we present the theoretical background for constructing DCF, drawn from the process-based approach which integrates many useful management concepts investigated separately. We discuss business process management as an integration foundation, quality and risk management, together with tax due diligence and financial control. In the fourth section, we expound on the method used to achieve our goal. The building process of DCF is based on the widely recognized Design Science Research method. In the fifth section, we examine the research results consisting of (1) the survey analysis of the accountants' attitude to SAF-T, mostly responsible for tax compliance, and (2) the design of DCF. The sixth section demonstrates the usefulness and evaluates the suitability of the constructed DCF. The discussion and conclusion (sections seven and eight) summarize our study.

THEORETICAL BACKGROUND

In the theoretical investigation, we build the background for constructing our framework integrating many useful management concepts (Figure 6), which are explored separately elsewhere. The term framework we use after Shapira (2000, p. 1314) where: (...) it (1) provides a structure to organize observations, and (2) describes the structure in a clear and precise manner. At the beginning of this section, we explore the process-based approach as an integration foundation. Subsequently, we outline other concepts, i.e. quality management, risk management, tax due diligence, and internal financial control.

Process management

Since the introduction of the business process reengineering concept outlined by Hammer (1990), Davenport (1993), and Hammer and Champy (1993), the idea of modeling the organization by means of processes has spread throughout academia and practice, building a foundation for a business process management discipline. Over almost three decades, it has matured, becoming a strategic means for constructing a competitive and resilient enterprise. However, the starting point of this idea can be found much earlier in Williams' (1967) paper where the author presented the concept of improving the production process efficiency by shortening its timespan and increasing the volume of production (5). The debate about modeling the organizations by processes was saturated in response to the publications by Davenport (1993), Hammer (1990) and Champy (1993). The hype generated by the promising results of business process reformulating convinced many practitioners to implement it. Studies concentrating on critical success factors caused by the radical change in the organizations reported some difficulties in determining the real outcome as a success or failure (Larsen & Myers, 1997). In the beginning, the financial indicators of the analyzed case displayed meaningful improvement, but later the implications were much more difficult. Rosemann (2014) explained this effect as a lack of a sufficient amount of methods supporting radical re-design and innovation in the organizational processes. The following decade showed a significant advancement in the theory and practice enriching the set of available process notations (e.g., Business Process Management Notation or Extension in Unified Modeling Language), process assessment tools (e.g., Six Sigma) and enterprise systems implementation (Sheer, 2000). Becker, Rosemann, and Uthmann (2002) presented some other fields where the process approach proved its value: lean management, activity-based costing, total quality management, process innovation, workflow management, and supply chain management. We would also like to extend this list with the capability maturity model (Van Looy, De Backer, & Poels, 2011) business continuity management, and enterprise information systems' implementation.

In the literature, we can find a variety of definitions presenting many different directions of approaches to business process management. Swenson and von Rosing (2015) published a review of approximately 100 papers and put forward their proposal...