Analysis of the influence of climatic conditions on the tax potential of Russian regions using cluster analysis

Gerasimova Anna Evgen'evna ORCID: 0000-0001-8480-6279 PhD in Economics Associate Professor; Department of Taxes and Tax Administration; Financial University under the Government of the Russian Federation 15 Verkhnyaya Maslovka str., Moscow, 127083, Russia kharitonova.ae@yandex.ru Other publications by this author

Introduction

The sustainable development of the country presupposes stability and balance of state and regional finances, counting on their systematic growth. Taxes are the main source of revenue for the budget at any level. The increase in tax revenues to the budget is facilitated by the development of entrepreneurial activity, therefore, the main task of the authorities is to create favorable conditions for increasing economic growth. However, the level of entrepreneurial activity directly depends on regional characteristics both in terms of climatic conditions and resource potential. As a result, it is important to correctly assess the tax potential of regions from the point of view of natural conditions in order to plan budgets at all levels.

The territory of Russia is so vast that it is located in various climatic zones - from the Arctic to the subtropical. Climatic features have a direct impact on the formation of the economic characteristics of the regions and determines the priority directions of their development.

An assessment of the tax potential of regions, taking into account climatic factors, will allow for a more complete assessment of tax revenues and determine to public authorities the potential amount of income necessary for the formation of the expenditure side of budgets at all levels and the competent allocation of budget funds. Therefore, the study of the tax potential of regions, as well as the factors that directly affect it, is an important aspect of macroeconomic analysis and acquires particular relevance and significance.

Literary review

Approaches to defining the concept of tax potential are interpreted in different ways depending on the authors. Thus, the simplest definition is given by F. S. Aguzarova and S. K. Tokareva: "The tax potential means the most probable and permissible amount of tax payments and fees in a particular economic space" ^[1] and Pechenskaya-Polishchuk M. A.: "the totality of the maximum possible tax revenues to the budget in the current economic conditions" ^[2]. These definitions do not disclose the conditions and factors that affect the tax potential.

The definition of G. L. Popova's tax potential is more complete: "this is the maximum possible amount of tax revenue to the consolidated budget of the region, taking into account the prevailing significant internal and external factors that determine the level of financial and economic development of the region" ^[3]. This definition also includes factors that can influence the level of development of regions, which more fully characterizes this concept, but does not disclose the factors themselves. At the same time, the following factors affecting the tax potential are highlighted in the work of V. V. Roshchupkina: regulatory, legislative, resource-raw materials, production, organizational, infrastructural ^[4]. A. A. Mironov identifies the following groups of factors: endogenous and exogenous, highlighting among them subjective and objective, historical and acquired ^[5].

A more complete definition is given by V. A. Shabashev and T. T. Malikaidarov: "The tax potential of a region is an economic category that represents the optimal amount of tax revenues to the budget in a certain territory operating in a given legal and economic field, with the most effective use of all available resources of a given territory in the conditions of the existing taxation system" ^[6]. The advantage of this definition is that the author characterizes the tax potential not as the highest possible level of taxes in the region, but as optimal when maximizing the use of all possible resources.

In the works of I. V. Maiburov ^[7], only three approaches to the essence of the concept of "tax potential" are distinguished: fiscal, inter-budgetary and resource. Much more approaches to definition have been revealed in the studies of A. S. Rogachev, M. P. Loginov and N. V. Usova ^[8]. The authors identify the following approaches: fiscal, or profitable, resource, institutional, mixed, or inter-budgetary, productive, reproductive, generating, transactional, systemic and investment approach and the method of synergetic dualism. Separately, a systematic approach to the definition developed by the authors should be highlighted: "the system of the existing and potential tax base of the territory, taking into account the negative effect of the regional shadow economy and the positive / negative influence of the state" ^[8]. This definition includes factors of the shadow economy, which are difficult to assess, but at the same time distort the real idea of the tax potential of the regions.

For this study, the tax potential of the region can be defined as the optimal level of tax revenues to the budget, taking into account regional peculiarities, while maximizing the effect of using all possible resources of the territory within the framework of laws and taking into account the existing tax system. At the same time, the potential maximum effect can be considered from the point of view of the current position of the region in the formation of budgets at all levels. Accordingly, the higher the tax revenues and the level of development of the region, the higher its tax potential.

The factors influencing the tax potential of the region are very diverse, they also include regional peculiarities, including the natural and climatic factor. Specialization and economic orientation depend on the meteorological features of the region, which directly affects the development of the region and, accordingly, the amount of taxes levied. The financial and credit mechanism of regional development is also based on climatic differences. Thus, in regions with a subtropical climate, agricultural production is primarily developed and preferential support mechanisms must be applied ^[9]. At the same time, there is a dependence that the higher the level of socio-economic development of the region, the higher the tax potential of the region and, accordingly, the higher the financial and economic security ^[10].

In a number of works, the grouping method is used to differentiate regions by tax potential. In ^[2], groups of regions were identified according to the share of tax revenues transferred to the federal budget from the volume collected. However, this indicator has a high variation over the years and grouping by only one indicator may not always be reliable when highlighting differences. The cluster analysis method allows not only to take data in dynamics, but also to use several indicators for a more reliable division, which makes differentiation more reliable and reliable.

In most literature sources, a set of factors for tax potential is assessed, while climatic conditions are not singled out separately. Thus, in the work of V. A. Shabashev and T. T. Malikaidarov ^[6], geographical and sectoral factors are highlighted. At the same time, groups of regions are distinguished: "developed mining and processing", "predominantly mining", "predominantly processing", "with relatively low shares of mining and processing industries". However, industry features are primarily influenced by natural and climatic conditions. Also, these groupings do not take into account regions specializing in agricultural production with a low tax burden. As a result, it is necessary to take into account not only the industry specifics, but also the climatic conditions on the basis of which they have developed. Madreimov A. O. ^[11] also notes the impact on the tax potential of location (the presence of borders with other countries) and the specifics of the regions. The geographical factor should also be taken into account.

The study by Mutascu M. ^[12] examines the hypothesis that climatic conditions determine the level of state tax revenues. At the same time, the dependence between tax revenues per capita and temperature is estimated by constructing an econometric model based on panel data. The disadvantage of this study is that only a factor characterizing temperatures is considered, but precipitation, which has not been taken into account, also plays an important role in fully characterizing climatic conditions.

The purpose of this work is to assess the impact of the climatic factor on the tax potential of regions by grouping using the cluster analysis method. The identification of common features in regions with similar climatic conditions of management will allow them to develop similar mechanisms for increasing their tax potential.

Materials and methods of research

The territory of our country is so large that the regions are located in different climatic zones - from the Arctic to the subtropical. Differences have a significant impact on the business environment. As a result, significant differences in meteorological conditions have an impact on the resource potential of the regions, the specialization of the regions and, accordingly, on the level of tax burden.

The meteorological features of the regions are unstable from year to year. Thus, the abnormal heat of 2010 led to a decrease in the value of gross crop production in prices of the previous year by 295 billion rubles (23.8%), which is more than 2 times the amount of state subsidies to agriculture ^[13]. For a more accurate assessment of the climatic conditions of the region, it is advisable to analyze the data in dynamics due to their high variation. The official collections of Rosstat annually publish data in the regional context on temperature and precipitation in July and January only. For a more complete assessment, it would be advisable to evaluate the data of the growing season, but they are not available in the regional context.

A better differentiation will allow for cluster analysis, which is based on the identification of groups based on several characteristics at once. The research methodology will consist of several stages.

Stage 1 – selection of factors characterizing climatic conditions in the regional context. At the same time, data cleaning is necessary because not all regions can be provided with information and normalization – to bring the data to uniform units of measurement. To differentiate regions by climatic conditions, let's take data for the last 3 years on the temperatures of January and July, as well as on the amount of precipitation in January and July.

Stage 2. Determination of the optimal number of clusters for the application of the k-means method. It is advisable to carry out this stage using the R programming language with built-in packages such as factoextra and cluster to automate the application of the elbow method. This method considers the nature of the variation in the spread of the total variation of the data with an increase in the number of groups k.

Stage 3. Conducting cluster analysis using the k-means method using the stats package. This method allows you to divide the population into k clusters, with each observation referring to that cluster, to the centroid which one it is closest to. At the same time, the selected regions that fall into the same cluster will be the most similar in terms of multidimensional principle.

Stage 4. Evaluation of selected clusters. The selected groups can be characterized in terms of climatic factors, which will reflect the quality of clustering. It is also necessary to assess the tax potential of regions by groups, depending on their climatic potential. It is also possible to assess differences in the level of agricultural development and competitiveness of groups of regions.

The application of this methodology will allow identifying groups of regions with similar climatic conditions in order to develop common methods and approaches to assessing the tax potential of regions.

The results of the study

The analysis was carried out in 79 regions of Russia (there are no data available for the rest of the subjects). Using the factoextra and cluster packages of the R programming language, the optimal number of clusters was determined using the elbow method, which considers the nature of the variation in the spread of the general variation with an increase in the number of k groups. According to the graph (Figure 1), it can be seen that 3 clusters are optimal, since the growth of k values is sharply reduced.

Figure 1 – Choosing the optimal number of clusters

Source: compiled by the author based on the results of the fviz_gap_stat function of the factoextra package

Cluster analysis was performed in the R programming environment using the stats package. The k-means method was chosen as one of the most popular methods of cluster analysis, which minimizes the dispersion of variances within groups and maximizes intergroup variation. The data was previously normalized due to differences in the measurement units of the indicators. The results are shown in Figure 2. Clustering of regions explains 42.7% of the data variation due to the large number of input variables (12) and a fairly high variation in temperatures and precipitation over the years.

Figure 2 – Results of cluster analysis in the R environment

Source: compiled by the author based on the results of the kmeans function of the cluster package

The composition of clusters is shown in Table 1. As you can see, the distribution of regions across clusters is uniform. The first cluster (group) included mainly the regions of central Russia, the eastern part of the country in group 2, and the southern regions in group 3

Table 1 – Cluster composition

Source: compiled by the author based on the results of cluster analysis

The division of regions into groups is shown in Figure 3.

Figure 3 –Distribution of clusters in the space of selected main components in the R programming environment

Source: compiled by the author based on the results of the fviz_cluster function of the factoextra package

In the figure, the main components are highlighted along the axes, which carry 44.2% and 18.2% variations of the initial data. The selected clusters are separated from each other and do not intersect, which confirms the quality of the separation. The distribution of regions in relative terms by climatic conditions is shown in table 2.

Table 2 – Characteristics of the climatic potential of the selected groups

Source: compiled by the author according to the Federal State Statistics Service

Table 2 shows that the regions with the most unfavorable climatic conditions fell into cluster 2. For all three years considered, the average annual temperature in July is lower than the national average by 2.3 C0, 3.2 C0 and 1.9 C0, respectively, for 2020-2022. Also, the average January temperatures of 9.1 C0, 10.6 C0 and 7.8 C0 are lower than the average for the years considered. According to the amount of precipitation in July, the regions of cluster 2 occupy an intermediate position and have the most stable character, this is evidenced by data from 2021, when precipitation in the regions of cluster 1 decreased sharply. At the same time, January has the least precipitation compared to regions 1 and 3 of clusters. According to the average score of climate productivity, as well as meteorological conditions, it can be said that in these regions, crop production cannot be strongly developed due to unfavorable conditions. At the same time, this group accounts for 84.4% of the country's area and 38.3% of agricultural land, mainly occupied by hayfields and pastures.

Cluster 3 includes regions with the most favorable climatic conditions. The average temperatures in July are above the values of cluster 1 and 3 and above the national average by 4.0 C0, 3.4 C0 and 2.6 C0, respectively. January temperatures are the highest among the groups under consideration - 5.0 C0, 7.5 C0 and 5.5 C0 above the national average, respectively. However, these regions are characterized by dry summer months, since July precipitation is the lowest in the groups under consideration, but the winter months are quite snowy (precipitation is higher than in the regions of cluster 2). Favorable conditions for agricultural production are confirmed by the average climate productivity score, which turned out to be 8.8 higher than the national average. However, this cluster accounts for only 7.2% of the country's area, while 41.1% of agricultural land.

The regions of cluster 1 occupy an intermediate position between groups 2 and 3. These regions are characterized by a temperate climate. However, the variation in values by year is the highest, since the average temperatures in July 2021 are 2.2 C0 and 1.8 C0 higher than 2020 and 2022, respectively, and the average precipitation in 2021 was 2 and 1.8 times lower, respectively, for the other groups such strong temperature and precipitation variations were not observed. These regions have the highest climate productivity score (140). The regions of this cluster account for only 8.4% of the country's territory, but at the same time 20.6% of all agricultural land. Agricultural production is well developed in the regions, but there may be risks associated with the climatic factor.

Table 3 – Characteristics of the tax potential of the selected groups of regions

Source: compiled by the author according to the data of the Federal State Statistics Service and the Federal Tax Service

According to table 3, it can be seen that the highest cash receipts from consolidated budgets are typical for the regions of cluster 2. For all three years considered, the indicators are on average 1 region higher than the national average by 8.6, 19.3 and 17.1 billion rubles, respectively. Due to the fact that the population of these regions is not large, the indicators per 1 person are more than 2 times higher than the values of regions 1 and 3 of the cluster for all 3 years considered. The situation is similar with budget expenditures: these regions are characterized by the highest expenditures both per 1 region and per 1 person. Also, these regions are characterized by the highest values of tax receipts, fees and other mandatory payments to the consolidated budget as of 01.01.2024. Thus, it should be noted that despite adverse weather conditions, the potential tax potential of the regions of cluster 2 is the highest. This is due to the fact that mining is developed in these regions, which is the industry with the highest tax burden (65.6% in 2023), while the tax burden of agriculture is only 4.5%.

The regions of the 3 clusters occupy an intermediate position in terms of revenues to the consolidated budget for an average of 1 region. However, per 1 person, the values become lower than those of the regions of the 1st group. This is due to the high population density of the regions of cluster 1, which includes the subjects of the central part of Russia. The situation with expenses is similar: on average, values for 1 region are higher than 1 cluster, and for 1 person they are lower. Thus, we can say that despite the development of agricultural production and the low tax burden of this industry ^[14] in the regions of cluster 3, the tax potential remains quite high.

The regions of cluster 1 have the lowest tax revenues. Despite the high population density, budget revenues on average for 1 region are the lowest in the country (below the average level by 11.3, 12.2 and 13.4 billion rubles, respectively, over the years). Also, these regions are characterized by the lowest expenses on average for 1 region (lower than the average by 9.4, 11.3 and 13.2 billion rubles).

Table 4 – Economic development of the selected groups of regions

Source: compiled by the author according to the Federal State Statistics Service

According to the overall level of economic development, regions of 2 clusters should also be distinguished. They are characterized by the highest gross regional product on average per 1 region, as well as per capita (25% and 45% higher than the average in Russia). The high level of development is confirmed by the largest number of allocated investments in fixed assets. These regions are characterized by a high level of industrial production, since these regions have the highest turnover of organizations (by 19.5% of the average level) and a balanced financial result (by 32.2%). At the same time, agriculture is practically not developed, and in terms of mining, the turnover of organizations is significantly higher than the average Russian level by almost 2.4 times, and in terms of the balanced financial result – by 2.3 times. As a result of the development of mining, these regions are the most promising in terms of tax potential.

The regions of group 3, despite the lowest values of the turnover of organizations and the balanced financial result, occupy an intermediate position among the considered groups in terms of gross regional product (6% higher than cluster 1). The picture is similar for investments in fixed assets per 1 region. However, on a per capita basis, the regions of group 1 exceed the values of group 3. It should also be noted the development of agricultural production in these regions (the turnover of organizations is above the average level of Russia by 18%, and according to the balanced financial result – 17.3%). Thus, it can be said that with an increase in the turnover of organizations in these regions, the tax potential of the regions of this group can be significantly increased.

The regions of cluster 1 are characterized by average indicators of turnover of organizations and balanced financial results relative to other groups. At the same time, agricultural production is quite developed in these regions, but mining is practically not developed. Despite the fact that this group mainly includes the regions of central Russia, the tax potential can be significantly increased.

Thus, it can be concluded that the tax potential of the region depends on specialization depending on climatic conditions. At the same time, the regions with the most unfavorable conditions in terms of climatic conditions have the highest tax potential, the regions of arid climate occupy an intermediate position, and the regions with a temperate climate have the lowest tax potential.

4 Discussion

The tax potential of the region plays an important role in the formation of budgets at all levels, which affects not only the economic development of the regions, but also the financial stability of the country. It is influenced by a huge number of factors, both internal and external to the region. For competent tax planning at the regional and national levels, it is necessary to take into account as much as possible all possible factors to improve the quality of forecasts, while also taking into account the climatic factor, including weather risks and anomalies.

This methodological approach to the analysis of the influence of the climatic factor on the tax potential of regions allows us to identify subjects with similar natural and climatic conditions and evaluate them together. At the same time, differences in the formed groups become noticeable both in economic conditions and in tax revenues and, accordingly, tax potential.

Similar conclusions about the high potential of regions with adverse weather conditions are confirmed in the study by Mutascu M. ^[11] noting that zones with moderate and low temperatures represent the best environment for tax revenues. It is also proposed to adjust the tax policy of countries taking into account climate maps. In this regard, the possibility of using cluster analysis for groupings of regions with similar business conditions will allow differentiating tax policy for similar groups of regions.

For our country, in order to increase the tax potential of regions for regions with similar climatic conditions, it is also necessary to develop a strategy for the development of those industries in which these groups specialize. Thus, for regions with unfavorable climatic conditions, special tax incentives can be developed for those organizations of the extractive industry whose budget revenues are in the top 10 for these groups of regions. This will increase the competitiveness of industries, production volumes and tax potential.

At the same time, the system of indicators for identifying homogeneous groups using cluster analysis can be expanded taking into account new factors, for example, taking into account the presence of borders with foreign countries, which will expand the methodological approach to analysis and will also allow identifying new features for similar regions.

5 Conclusion

As a result of the conducted research, the dependence of the tax potential of the region on climatic conditions has been revealed. At the same time, the regions of the eastern part of Russia with the most severe climatic conditions have the greatest weight in budget revenues at all levels. This is primarily due to the specialization of the regions in industries with a high tax burden.

Regions with the most favorable economic conditions, despite their specialization in the production of crop production and preferential tax conditions for agricultural producers, have high tax revenues and a fairly high tax potential.

The regions of the central part of Russia with moderate climatic conditions have the least role in budget revenues, however, despite this, when developing government measures to support small and medium-sized businesses, they can increase their tax potential.

The developed approach to the differentiation of regions by climatic factor will allow applying measures not for each individual region, but for homogeneous groups of regions, which will reduce the complexity of work and increase the tax potential. To increase the tax potential, a set of measures can be implemented to improve tax policy and tax administration in order to simplify the procedure for paying taxes for the public and businesses ^[15]. At the same time, according to a number of authors, it is necessary to provide for "intensive social, economic and environmental development of an administrative-territorial unit, an effectively functioning tax system, taking into account compliance with the principle of stability of tax legislation" ^[16]. At the same time, for qualitative forecasting of tax potential, it is necessary to take into account a set of measures, which requires digitalization of all flows with further automation of analysis based on the introduction of the latest technologies for processing large amounts of data ^{[17, 18]}. And for the productive optimization of the tax burden by an economic entity, it is necessary to take into account all the characteristics of the activity and features of the industry ^[19].

To increase the tax potential of the regions, it is necessary to take into account all factors, monitor the ongoing processes, and analyze to develop comprehensive measures that will contribute to the growth of revenues to budgets and improve the economic situation in the region.