By CPA Nyakoi Oreko Godfrey.
Technological evolution and related innovation have created a paradigm shift in the management of tax administration in many authorities. Tax authorities are moving towards embracing a more integrated digital tax administration away from the conventional models of tax administration. Taxpayers are significant players in tax administration matters, since they provide data to the tax administration authorities regarding their taxes (VAT, Corporate income tax, etc.).
To this extent, taxpayers are the first-party source of tax data. In addition, they are regarded as third-party data providers by providing tax information about individuals (employees) and other corporate (customers) to the tax administration agencies. Currently, tax agents collect raw data on personal or business incomes as much as possible. The reason behind the collection of data is that the more data is available, the easier it is to detect discrepancies. If there are no discrepancies detected, it is more likely that the data provided is accurate and correct. Data quality is defined by the absence of any detected discrepancies. In the era of technological evolution, data discrepancy detection has been made easier and faster.
Technology has enabled tax authorities globally to adopt the emerging technology tools to effectively and
efficiently manage tax administration. Among the emerging technology tools challenges is how to select among the various combination of technologies to give an optimal output. When choosing technologies to adopt, tax administrators ought to consider the already established systems, quality of data available, technical skills of the employees, and the capacity to use technological innovations. The adoption of digitalization tools like cloud computing, big data and data analytics, artificial intelligence (AI),
When choosing technologies to adopt, tax administrators ought to consider they already established systems, quality of data available, technical skills of the employees, and the capacity to use technological innovations.
Blockchain, quantum computing, and the Internet of things (IoT) has provided a range of new possibilities such as process optimization, and real-time identification of risks and automation of processes. Globally, various tax authorities are at different stages of the digitalization journey. The following are some of the current technologies that may be used by tax agents to efficiently manage tax administration.
Table 1; Technologies and their practical application for tax purposes
| Technology tool
| Description | Application for tax purpose |
| Cloud Computing Big Data & Data Analytics Artificial Intelligence (AI) Blockchain | This is a technology used for shared storage, computational capacity, and application software provided externally and interconnected by the internet. Cloud computing allows for the remote delivery of on-demand computing services over a network, commonly pay-for-use basis. Examples of cloud service providers include Oracle, SAP, and Microsoft (Azure) | This technology has the capacity of reducing costs and increasing institutional agility by allowing tax agents not solely depend on IT equipment like specific computer hardware, and data centers among others. This is because cloud computing does not depend on a certain specific equipment band and has the advantage of being accessed from different locations |
| Big Data refers to the concept of huge volume, variety, velocity, veracity, and the value of the data while data analytics refers to the independent evaluation of data or content by employing statistical software or tools to reveal more data patterns, make forecasts or generate reports for decision making.
AI is a significant modern technology that uses the application of advanced analysis and logic techniques by the use of machine learning to interpret events and support automated decisions.
| Tax authorities have access to the huge amount of data collected through tax returns, assessments, tax collections, etc. Big data and data analytics become very useful in processing large data. Most tax agents find it difficult to process the immense amount of data collected. Some tax agents employ risk analysis and sampling of relevant data by the use of advanced analytic methods like machine learning and Artificial Intelligence (AI). These methods are significant in case of a massive amount of data because they allow faster and more accurate analysis To optimize the use of Big data and AI, tax authorities need to significantly invest in tools and skills and put together an IT-skilled tax team, which includes professionals with different backgrounds like economists and statisticians. In addition, tax agents should use other advanced technology tools in analyzing huge data like network analysis, text mining, and web scraping. | |
| Blockchain is a system of capturing data in a manner that makes it almost impossible to alter or hack the system. It is a digital ledger of transactions duplicated and distributed across the entire network of computer systems on the blockchain. | This is a major technology that can disrupt how tax systems are managed. Currently, it is being embraced by developed economies as a way of modernizing the existing tax systems. The technology has numerous potential applications both in the business world and tax administration by eliminating the use of third parties with the capability of allowing real-time data sharing, tax collection, and administration. However, implementing this technology requires harmonizing the various underlying principles nationally as well as internationally. |
As shown in the above summary, technology tools can be applied in various ways to deliver better and
more efficient services to taxpayers through e-services. The technology tools can also be used in analyzing the huge amount of data collected by tax authorities by the use of data analytics technologies like Big Data, which will help tax authorities efficiently manage risks associated with tax avoidance and frauds. The application of technology tools in tax management will also profit various government agencies like Kenya. Revenue Authority, EACC, OAG, and ODPP, since it will facilitate efficient cooperation and sharing of information about various taxpayers be it individual or corporate.
technology tools can be applied in various ways to deliver better and more efficient services to taxpayers through e-services. The technology tools can also be used in analyzing the huge amount of data collected by tax authorities by the use of data analytics technologies like Big Data, which will help tax authorities efficiently manage risks associated with tax avoidance and frauds. The application of technology
tools in tax management will also profit various government agencies like Kenya Revenue Authority, EACC, OAG, and ODPP, since it will facilitate efficient cooperation and sharing of information about various taxpayers be it individual or corporate.
Challenges underlying new technology tools in tax administration.
Despite the new technology tools offering diverse application opportunities in managing and transforming tax processes and making them efficient and effective, it equally poses risks in the implementation of the new technology. The following are some of the risks associated with the application of the new technology tools:
Quality:
Mishandling of data at the raw data level signifies that there will be an inherent risk of making errors and omissions, which may lead to an inaccurate representation of the taxpayer’s financial position. Inherent risk increases especially when huge and heterogeneous data items are involved, which makes it difficult to coherently model it in a similar database. A perfect example is a financial information received from foreign countries in the context of automatic exchange of data, which needs to be marched against the details of the taxpayer’s record to correctly identify the taxpayer to which each data item refers.
Privacy:
When dealing with huge massive data, data protection becomes paramount. Data protection does not only relate to unauthorized access to data but also violations of the data protection regulation in handling and usage of the accessed data. Always data protection regulations should be prioritized when massive data is involved, including in situations subject to strict ethical commitments like tax administration. For example, when analyzing invoicing data may reveal sensitive information about the taxpayer like buying behavior. On the contrary, blocking access to some part of the information for the sake of privacy can exclude crucial characteristics to be considered in the overall picture. This situation has the potential of increasing the risk of incorrect predictions and false positives.
Security:
Under all circumstances, observing personal data integrity becomes paramount in the wake of embracing technology. This is particularly crucial when dealing with massive data treatments. Data access security must be guaranteed at the IT infrastructure level, autonomously of how data is used in the actual sense. Due to their nature of work, data scientists, are usually expected to migrate masses of data, for instance, when migrating data preparation from one database platform to another, or from one workstation to the other. This has the potential of leaking data outside the required boundaries of a safe IT environment, which potentially can increase the risk of unauthorized data access.
when analyzing invoicing data may reveal sensitive information about the taxpayer like buying behavior. On the contrary, blocking access to some part of the information for the sake of privacy can exclude crucial characteristics to be considered in the overall picture. This situation has the potential of increasing the risk of incorrect predictions and false positives.
Algorithms:
The system application of machine learning algorithms to a huge amount of data has the potential of introducing new risks associated with the effects of automated processes. Powerful algorithms like neural networks can detect relationships among the data that are impossible to detect when other methods are employed. These types of algorithms are referred to as black-box algorithms because they usually do not give particulars about the motivations that led to a specific outcome. For instance, when employing machine learning techniques in tax administration, said the selection of audit sample, interpretability is a desirable feature because auditors need to be clear about the motivation for them to begin an audit exercise.
Therefore, a lack of interpretation of automated processing is also risky in itself because it has the potential of hiding algorithmic bias, which may result in a systematically unbalanced result towards a certain part
of the population, usually derived from errors acquired in the training phase like incorrect assumptions.
Conclusion:
Technology can be applied in core functions in tax administration ranging from taxpayer, registration, filing, and declaration of tax payments, audits, taxpayer objections and appeals, enforcement measures, tax fraud investigations, and debt management Despite the risks to the algorithm, privacy, security, and quality of the
data, technology can be employed to facilitate cooperation between various government agencies for efficient and effective tax administration. Since there is exponential growth and relevance of digital platforms and e-commerce globally, there is a paradigm shift in the role of digital platforms in the collection
of value-added tax and income taxes on online sales.
