Tax administration in a changed reality: how to involve taxpayers in the digital environment ?

Anisimova Anastasiia PhD in Economics Postgraduate student, Department of Accounting and Taxation, Plekhanov Russian University of Economics; Scientific Associate, Financial Research Institute of the Ministry of Finance of the Russian Federation 127006, Russia, Moskva, g. Moscow, per. Nastas'inskii, 3 str. 2 a.a.anisimova@icloud.com Other publications by this author

Introduction

The relevance of the topic is due to the significant role of digital technologies in modern society. It is no exaggeration to say that at present the processes of digitalization are widespread all over the world, in all sectors of the economy. The differences between countries are only in the degree of digitalization development. In Russia, the digital society is developed at a high level, including due to the accessibility of the Internet and devices for citizens, such as PCs, laptops, smartphones, tablet computers, allowing interaction within the digital environment.

In the Russian Federation, digital technologies are currently developing very rapidly in tax administration. As a result of the ongoing digital processes in tax administration, there have been changes in the interaction of tax authorities with taxpayers. The approach has become more client-oriented. The taxpayer is now a customer, a recipient of digital tax services. The quality of the services provided largely depends on whether citizens will be able to interact in the digital tax environment in the future. At the same time, the tax authorities are offering more and more new electronic services to taxpayers, including in the framework of smartphone applications, which makes them especially convenient for use anywhere in the country with the Internet.

In this regard, the purpose of this publication is to assess the digitalization of tax administration in Russia.

In the course of the study , the following tasks were solved:

- identify the subjects of the Russian Federation where there are barriers to the introduction of digital technologies;

- to present on a practical example the identification of the most discussed topics of taxpayers in the framework of monitoring the Internet space;

- to show the results of digitalization of tax administration to increase the collection of tax payments to the budget system of the Russian Federation.

Development of digital interaction of taxpayers with tax authorities in the world

The role of tax services in terms of interaction with the taxpayer in the world has changed over the past decade. The tax authorities are currently not only implementing the state tax policy, but are also the entry point for large sets of financial data and the first stage of interaction between the citizen and the government. The digital processes laid down during this time period were tested during the COVID-19 pandemic, when governments increasingly relied on tax authorities in two important aspects: ensuring budget revenues and interaction between the government and taxpayers during the deployment of social support for the population. The digital transformation of tax administration at the present time determines the goals that tax services set for themselves at the present stage^[1].

Thus, in the OECD document "Tax Administration 3.0", which was published by the OECD Forum on Tax Administration (FTA), the following main objectives of tax administration were highlighted:

- unhindered, highly efficient revenue collection through optimized and automatic administration of the tax system, covering a broader tax base, which includes the use of data-based approaches that minimize gaps in the collection of tax payments without necessarily raising tax rates or adding new tax regimes;

- increasing the level of data transparency, security and trust through the use of digital platforms to manage the main processes when interacting with taxpayers, such as tax registration, filing documents, payments and dispute resolution, in order to make them understandable to taxpayers;

- minimizing the compliance burden through the introduction of digitized, optimized direct submission of documents to tax authorities by the taxpayer, as well as their processing and evaluation;

- improving the efficiency of tax administration by rethinking the process of interaction with business, especially when providing data on compliance with requirements, using technologies to conduct, accelerate and optimize decision-making processes during inspections of organizations by tax authorities, using artificial intelligence and machine learning technologies;

- providing a link between taxpayers and government programs (for example, when deploying economic incentive packages) and turning tax authorities into a government data bank; and

- the use of advanced taxpayer data for analytics (for example, to track payments and the use of money transfers, monitor the consumption of goods and simulate the reaction of taxpayers to the introduction of new taxes, such as, for example, carbon taxes)^[2].

The number 3.0 in the title of the document means the version number of the digital transformation of tax administration. A similar approach can be seen in the OECD documents in 2020. It means the following gradation of stages of development of tax administration in the conditions of digitalization:

- "Tax Administration 1.0" is a traditional paper form filled out manually and submitted by mail or in person.

- "Tax Administration 2.0" is digitized electronic versions of forms and procedures 1.0, such as PDF or online tax forms posted on the websites of tax authorities.

- "Tax Administration 3.0" is a stage that represents an emerging wave of digital and automated tax processes that are embedded in the same systems that taxpayers use to carry out their taxable activities.

In early studies of digitalization processes in the tax administration of the OECD, much attention is paid to the issues of interaction between taxpayers and tax authorities in the digital environment. The ideas laid in the foundation of such studies have largely helped to see the transition from "Tax Administration 2.0" to the next stage. The main idea is that a taxpayer who has once used the digital environment to solve his tasks within the framework of submitting reports or paying tax payments needs to provide all opportunities for further interaction within the online space in order to exclude traditional channels of information transmission, such as personal visits to tax authorities or visits to specialized centers (MFC), mail, telephone, etc. The taxpayer's retention in the digital environment is carried out through its formation in such a way that all problems and difficulties that arise are solved through digital tools. For example, if it is not possible to carry out any reporting operation within the framework of the provided interface, then there is a pop-up window nearby that you can use to send a message or make an online call. Moreover, tax authorities can monitor the online environment (social networks, forums, etc.) to identify emerging problems and determine the demand for digital services using tools such as Hootsuite, TweetDeck and Sprout Social, which provide the ability to search by "keywords" (keywords can be such as a declaration of income tax or tax refund). Working with such tools allows you to see the topics of discussions in the field of taxes and current problems. In the future, this information can be used to form a tax policy in the field of digital services development ^[1].

         Some aspects of digital interaction between tax authorities and taxpayers, which could be seen in earlier OECD documents, where the digitalization of tax administration was considered, may be revised in the future. As an example, voluntary compliance with tax legislation can be cited. Although at the beginning of the development of digital processes in the OECD documents, since 2007 it was possible to find a tendency to stimulate voluntary compliance by simplifying the interaction between taxpayers and tax authorities. That is, it was assumed that if the costs of compliance with tax legislation were reduced due to digitalization for taxpayers, the level of voluntary compliance would increase. Although taxation is not voluntary, the concept of "voluntary compliance" has been widely used because in many parts of the current tax system, taxpayers make choices about reporting, calculating and paying taxes. This choice is not only about compliance or non-compliance, but also covers the choice regarding the efforts made to do everything right, regarding the maintenance of documentation, taking time to fill out forms correctly, eliminating any misunderstanding and meeting reporting requirements and deadlines. However, currently in the OECD document "Tax Administration 3.0" voluntary compliance is already presented as a structural limitation of the tax administration system, which can lead to permanent problems of compliance with tax requirements and collection of payments. Despite the fact that at the present stage of development, tax authorities are making great efforts to support voluntary compliance, including through the development of new digital services, new communication channels and the development of approaches more focused on taxpayers, where there are options for compliance, some of the choices made will inevitably lead to non-payment of significant amounts of taxes, whether intentionally or as a result of making mistakes^[3]. Perhaps this is due to a change in the approach to voluntary compliance with legislation in the OECD documents. However, in some cases, voluntary compliance still remains relevant and a number of digital solutions aimed at improving the interaction of tax authorities with taxpayers have proved successful in terms of increasing tax collection. Here we can give an example from the Russian practice of the new tax regime "Professional Income Tax" and a mobile application related directly to this tax regime. The purpose of the development of this tax regime was largely to bring micro entrepreneurs out of the shadows. Under this tax regime, there are no reports and declarations, and the receipt is generated directly in the "My Tax" application (there is no need to install a CCT). The entrepreneur, in fact, decides for himself what income he should indicate in the application. That is, such a regime is largely aimed at the honesty and integrity of economic agents. The approach based on voluntary compliance with legislation is more relevant here than the exercise of control, because we are talking about micro business. The implementation of strict control here, as representatives of the tax authorities have repeatedly commented, will be more expensive in terms of costs.

Although digitalization has many advantages, it is not always without problems that are barriers to digital transformation. Some problems are temporary, such as poor change management; some can be worked on, such as taxpayer identification, but problems such as digital exclusion will remain forever.

Digital exclusion means that some taxpayers, due to their inability or unwillingness, do not use digital methods to interact with the tax authority. There are many reasons for this, such as lack of digital skills, distrust of online interaction, remoteness from reliable Internet infrastructure, poverty or disability. Many of the benefits of the digitalized tax ecosystem will be negated if a large number of taxpayers continue to use paper, personal, telephone or other old systems to interact with tax authorities. These methods are much more expensive per taxpayer. Significant activity outside the digital space also undermines the accuracy of data analysis and reduces the effectiveness of tax audits. Many tax authorities have sought to accelerate the introduction of digital methods of reporting and payments, making them mandatory. However, for people deprived of access to digital technologies, such an approach can be painful. Tax authorities in some countries have organized the process so that taxpayers pay for assistance in filing an application if they cannot do it themselves. This raises the question of fairness, for example, for applicants with certain disabilities who may not be able to comply digitally through no fault of their own. Other countries have made electronic filing optional, but have provided incentives, such as later payment dates or discounts, for those who choose to use it. Still others have made digital interaction mandatory, but with exceptions.

Tax authorities also need to consider different options for different groups of people excluded from the digital environment. Those who do not have the financial means or computer knowledge to submit documents electronically can be supported by incentives to purchase computer equipment or digital skills courses. Taxpayers can be informed about how taxes are collected and what they are spent on. This will help to gain the trust of those who do not submit documents digitally due to doubts about the tax system. Those who do not have reliable Internet access can be connected using an extended network. But some reasons for excluding taxpayers from the digital environment, such as disability, mean that it is impossible to include all taxpayers in one system. That is, there is no complete solution to the problem of digital isolation, and therefore it remains a key barrier to any digital development program^[4].

Assessment of digitalization of tax administration in Russia

Identifying barriers to digitalization among the Russian population, it is necessary to take into account regional peculiarities. Currently, there are 89 subjects in the Russian Federation and the population of each of these regions perceives modern digital technologies differently. In this regard, there is even a concept - the digital divide of Russian regions. The digital divide of regions is understood as the differentiation of different segments of the population according to the degree of accessibility of digital infrastructure and the level of information culture^[5]. The concept of the "digital divide" in relation to Russian subjects has been around for a long time. At the same time, the process of digitalization is undergoing changes (tax administration 2.0 is fundamentally different from tax administration 3.0), which means that the content of the concept of "digital divide of regions" is also changing. Currently, the digitalization process (tax administration 3.0) differs from earlier stages in that, with all the clarity and transparency of the digitalization program, it is not easy to implement it, and one cannot count on an immediate result, since investments in infrastructure projects to expand online interaction no longer give a quick and easily measurable effect as it used to be. That is, it is now much more difficult to bridge the digital divide. This is because the current digital divide in Russian regions is more determined by demand, that is, by the needs and requests of the population of the subjects of the Russian Federation, than by supply, which implies the services and services of suppliers and providers. The standard of living in Russian regions varies, and therefore the needs for digital services among the population also differ. In this regard, tax authorities, without losing sight of the development of digital supply (by conducting experiments offering consumers various models of digital interaction, and ensuring the competitiveness of products), need to develop digital demand ^[2].

If the digital interaction of the population, that is, individuals, with the tax service is currently completely voluntary and tax authorities are largely interested in increasing demand for electronic tax services, then the requirements for organizations are more stringent. Although since the advent of electronic document management (EDI), companies have switched to it voluntarily, the state is expanding the list of organizations that need EDI to work every year. For example, starting from 2022, electronic document management had to be used by organizations to provide accounting reports. The Federal Tax Service has developed the concept of electronic document management (EDI), according to which 95% of all invoices (not classified as state secrets) in Russia, as well as 70% of transport and consignment notes will be electronic by the end of 2024^[6].

Having data on the information society in the Russian Federation, it is possible to analyze the problem of digital exclusion of the population. For example, by determining which regions of the country are the closest in terms of digital development in order to identify barriers to the introduction of new technologies, including such as digital exclusion. For this purpose, let's consider three sets of data (for 2019) on the use of digital technologies by the population and organizations:

- the population interacting with state authorities and local self-government, according to the methods of interaction in the subjects of the Russian Federation;

- factors constraining the use of the Internet by the population in the subjects of the Russian Federation;

- organizations using electronic data exchange technologies and automatic object identification (RFID) technologies in the subjects of the Russian Federation^[7].

It is possible to carry out clustering of subjects by the k-means method. Clustering will be performed for two data sets:

- the population interacting with public authorities and local self-government, according to the methods of interaction in the subjects of the Russian Federation;

- organizations using electronic data exchange technologies and automatic object identification (RFID) technologies in the constituent entities of the Russian Federation.

The data set "factors constraining the use of the Internet by the population in the subjects of the Russian Federation" has a lot of gaps, so clustering does not make sense.  

First you need to perform data normalization. We subtract the mean from each attribute and divide by s, where s is the standard deviation of the attribute.

,              

Using the elbow method, we will determine how many clusters we need to take. In two cases, there are three clusters.

In the dataset "Population interacting with state and local government bodies, according to the methods of interaction in the subjects of the Russian Federation" clustering was carried out on two grounds (interaction via the Internet was not considered when creating clusters):

- personal visit;

- visit to the MFC.

As a result, one of the three clusters included only four subjects of the Russian Federation – those in which the values of the two indicators are the highest: the Kursk Region, the Kurgan Region, the Yamalo-Nenets Autonomous Okrug and the Altai Republic. That is, these are the regions where the population mainly solves their issues through visits to government agencies and MFC.

It is possible to consider in the second set of data the reasons why the population in these four regions uses the Internet to a much lesser extent compared to other subjects of the Russian Federation. In the Yamalo-Nenets Autonomous Okrug, this is a lack of necessity (unwillingness to use, no interest), as well as a lack of skills to work on the Internet and high costs of connecting to the Internet. In the Kurgan region and the Altai Republic, the main reason is the lack of necessity (unwillingness to use, no interest). In the Kursk region, the lack of necessity (unwillingness to use, no interest) and lack of skills to work on the Internet can be called as a barrier to using the Internet.

Thus, it can be seen that the main reason why the population does not use the Internet according to Rosstat is the lack of necessity, which also includes unwillingness to use and lack of interest among the population. It is worth noting that this is only a general picture, and it is necessary to examine each of the four regions listed separately in order to understand why it is easier for the population to solve their tasks in these subjects within the framework of interaction with public authorities not via the Internet, but through personal visits.

In the dataset "Organizations using electronic data exchange technologies and automatic object identification (RFID) technologies in the subjects of the Russian Federation", the following features were identified, according to which clustering was carried out:

- electronic data exchange between internal and external information systems;

- automatic object identification (RFID) technologies.

Automatic object identification (RFID) technology is in demand where control is required when moving objects. For example, in production (accounting for raw materials), in a warehouse (protection against theft and theft of products), in retail sales (tracking goods at the stages from the manufacturer to the counter) or in a library (issuing books, preventing theft).

         There are also three clusters here. Nine subjects of the Russian Federation are included in the cluster where two signs have the least importance: Tver Region, Nenets Autonomous Okrug, Republic of Kalmykia, Republic of Dagestan, Kabardino-Balkarian Republic, Republic of North Ossetia–Alania, Chechen Republic, Kurgan Region, Republic of Sakha (Yakutia). That is, with regard to organizations, the problem of digital exclusion in the field of tax administration should be paid attention in these nine subjects.

         The ideal way for the tax system to work is when all issues are resolved through a digital environment, even problems with the online channels of interaction themselves when submitting reports or paying taxes from individuals and organizations. The existing world experience shows that tax authorities also need to monitor taxpayers' problems on specialized forums or on social networks. Indeed, taxpayers are more likely to try to solve the problem without contacting the tax authorities, asking for advice on their difficulties in the Internet space. The analysis of such messages by the tax service can help solve the problem for a large number of taxpayers at once. Modern tools for working with big data allow for such an analysis. Here you can use a simplified example to briefly illustrate how it can be implemented. We will talk about the semantic analysis of the text, which was previously taken using python tools and its libraries from one of the tax forums where tax problems are discussed. For example, the topics of messages from the last 10 tabs (200 messages) were taken from the site. Then the text of the messages received from the forum was processed:

- we reduce the entire text to lowercase;

- remove service characters (punctuation, spaces, line breaks, etc.);

- delete stop words (conjunctions, prepositions, other semantically irrelevant words)

- we bring all words to normal form (lemmatization)

As a result, we get the phrase "6 personal income tax half-year" instead of "6-personal income tax for half-year". You can see that after processing the message text, all words are reduced to lowercase, and there is also no hyphen and preposition. One more message can be given to see an example of lemmatization. So, we get "leave to care for a child of 1 5 years" instead of "leave to care for a child up to 1.5 years". Next, we will convert all messages into TF-IDF vectors and cluster messages using the k-means method. As a result, all messages on topics on the forum were divided into three groups. The first two groups included messages related to the discussion of personal income tax, the second group - messages on the topic of VAT, and the third – all other messages. From which it can be concluded that the most significant topics for taxpayers are: personal income tax and VAT. Having identified the most difficult topics, then you can continue such an analysis directly within these topics, which will help determine the real problems faced by organizations in terms of personal income tax and VAT.

The following can be said about the current stage of digitalization of tax administration in Russia. Individuals and organizations are already heavily involved in the digital tax environment. It is safe to say that the digitalization of tax administration in Russia already has many components of stage 3.0. The Federal Tax Service of Russia offers an extensive list of various digital services on its official website: personal accounts, support measures, business registration, information from registers, tax calculators, tax accounting, electronic document management, international taxation, software, reference information, feedback^[8]. Most of the electronic services presented on the website of the Federal Tax Service of Russia are also presented in convenient mobile applications for both individuals and entrepreneurs. In this regard, it would be interesting to make calculations to find out whether there is any sense from the ongoing digitalization measures in terms of increasing budget revenues.

The effect of digitalization of tax administration can be checked using statistical tests. To do this, in the FIRA PRO program, we will select large (7000 organizations) and medium-sized companies (12,000 organizations) and conduct tests to compare the amount of taxes collected for the two pre-covid years 2018 and 2019. Python and its necessary libraries (Pandas, Numpy, SciPy) were used directly for statistical calculations. Why are these particular tax periods being compared? During these years, online CCTS were actively put into operation. Moreover, in 2019, the list of companies for which the use of online cash registers has become a necessity has been significantly expanded compared to 2018. Obviously, there will be an assumption that other factors not related to the digitalization of tax administration are not taken into account. However, if we consider the media for that period of time, the main agenda was the introduction of online cash registers and all other factors affecting the increase in tax collection remained insignificant. In 2019, online CCTS were introduced for the entire service sector, including organizations and individual entrepreneurs who are both on the main and simplified taxation system. Micro-entrepreneurs were also affected by digital innovations in 2019 regarding the introduction of a new experimental tax regime "Tax on professional income". However, the transition to this regime and coming out of the shadows was a voluntary matter, and therefore medium and large enterprises will be considered here. It was much easier to collect statistics on these companies. So, first let's consider whether there is a statistical effect in two samples – the VAT paid by companies in 2019 and 2018. First, let's make a calculation for medium-sized companies. It makes sense to apply the Mann-Whitney test, since the distribution is not normal, as shown by the Shapiro test. Calculations have shown that the null hypothesis is rejected. A statistically significant effect takes place (p-value – 0.000). That is, the samples are not equal and in 2019 VAT was collected in a larger volume compared to 2018, as the median values of the two samples tell us. We carry out the same procedure for large enterprises and get the absence of a statistically significant effect (p-value – 0.071). This can be explained by the fact that for large companies there are other ways to engage in digital interaction with tax authorities. For example, tax monitoring. Procedures in 2018 and 2019 within the framework of digitalization of tax administration by tax authorities were carried out mainly for medium and small enterprises in the service sector. In the case of income tax, the same result was obtained. For medium-sized enterprises, there is an effect (p–value - 0.004, that is, the null hypothesis of the Mann-Whitney test is rejected, the samples differ), for large enterprises there is no effect (p–value - 0.131, that is, the null hypothesis of the test is not rejected).

Thus, within the framework of the study, the following results were obtained:

1. The regions of the country closest to the level of digital development and barriers to the introduction of new technologies are identified. The regions where the population mainly uses traditional channels of communication with tax authorities include the following: Kursk Region, Kurgan Region, Yamalo-Nenets Autonomous Okrug and the Altai Republic. The main obstacle for these four subjects of the Russian Federation for the use of digital technologies by individuals is the lack of necessity (unwillingness to use, no interest). The second most important is the lack of skills to work on the Internet, and for the Yamalo-Nenets Autonomous Okrug, the high costs of connecting to the Internet. In general, more detailed studies are required to understand the emergence of barriers to the development of the digital environment, since the overwhelming number of regions of Russia, with only one exception (Khanty-Mansi Autonomous Okrug – Yugra), the prevailing deterrent is the lack of necessity (unwillingness to use, no interest). But there is a big difference between lack of necessity and lack of interest. Unfortunately, the existing monitoring of the digital environment in Russia does not take into account these details.

The regions where organizations are less involved in digital electronic document management technologies include the following nine: Tver Region, Nenets Autonomous Okrug, Republic of Kalmykia, Republic of Dagestan, Kabardino-Balkarian Republic, Republic of North Ossetia–Alania, Chechen Republic, Kurgan Region, Republic of Sakha (Yakutia).

2. By conducting a semantic analysis of the texts of messages from one of the tax forums, the most discussed topics (personal income tax and VAT) of taxpayers in the Internet environment were identified.

3. The results of digitalization of tax administration to increase the collection of tax payments to the budget system of the Russian Federation are shown. Statistical tests were conducted for large (7000 organizations) and medium-sized companies (12,000 organizations) to compare the amount of taxes collected (income tax and VAT) for 2018 and 2019. For large companies, the test did not show a difference in the two samples, and in the case of the average, a statistical effect took place. Medium-sized companies paid more taxes in 2019 than in 2018.

^[1] Launching a Digital Tax Administration Transformation: What You Need to Know / Asian Development Bank: https://www.adb.org/publications/digital-tax-administration-transformation

^[2] Launching a Digital Tax Administration Transformation: What You Need to Know / Asian Development Bank: https://www.adb.org/publications/digital-tax-administration-transformation, Tax Administration 3.0: The Digital Transformation of Tax Administration / OECD:

https://www.oecd.org/tax/forum-on-tax-administration/publications-and-products/tax-administration-3-0-the-digital-transformation-of-tax-administration.pdf

^[3] Tax Administration 3.0: The Digital Transformation of Tax Administration / OECD:

https://www.oecd.org/tax/forum-on-tax-administration/publications-and-products/tax-administration-3-0-the-digital-transformation-of-tax-administration.pdf

^[4] Digitalisation of Tax: International Perspectives / ICAEW: https://www.icaew.com/-/media/corporate/files/technical/technology/thought-leadership/digitalisation-of-tax.ashx

^[5] Ratner S.V. Digital divide of Russian regions as a threat to socio-economic development. Countries // Threats and Security 39(180) 2012

^[6] When EDO will become mandatory / The outline of the Diadems: https://kontur.ru/diadoc/spravka/25112-kogda_edo_stanet_obyazatelnym#header_25112_1 , The concept of development of electronic document management in economic activity / Federal Tax Service of Russia: https://www.nalog.gov.ru/rn77/related_activities/el_doc/el_bus_entities / ^[7] Information Society in the Russian Federation / Rosstat:

https://rosstat.gov.ru/folder/210/document/13251

^[8] All services / Federal Tax Service of Russia: https://www.nalog.gov.ru/rn77/about_fts/el_usl/