“An integrated approach to assessing the level of fiscal policy decentralization”

The main purpose of this study is to introduce an integrated approach to the methodol- ogy of assessing the level of fiscal policy decentralization. It is proposed to evaluate the fiscal policy decentralization of the state according to three functional components: decentralization of the process of local budget revenues formation (includes five indi-cators); decentralization of local budget structure (includes six indicators); decentral- ization of intergovernmental budgetary relations (includes five indicators). The expe-diency of forming an integral indicator of the level of fiscal policy decentralization as the geometric mean of three sub-indexes formed by its main functional components is substantiated. It has been proved that the level of fiscal decentralization in Ukraine decreased at the end of 2017, compared to 2004, but was medium with acceptable risks of fiscal policy modernization. Instead, in 2014, the lowest numerical value of the decentralization level was recorded, which corresponded to the critical level of the integral indicator with significant obstacles to the modernization of fiscal policy. The results obtained confirm the feasibility of implementing the decentralization reform in Ukraine, which started in 2014, and demonstrate its effectiveness.


INTRODUCTION
Decentralization processes create the preconditions for changing fiscal policy priorities towards the successful modernization of its components at the local level. Local governments are involved in the implementation of fiscal policies, which enhances their accountability in fulfilling the delegated obligations. The decentralization reform, which started in Ukraine since 2014, has become a key area of development of domestic public administration, which is implemented in order to transfer the power and financial powers of the central government to the benefit of local self-government. Procedurally, territorial-administrative units are given the extension of powers regarding the independence and self-government of decision-making and the exercise of delegated powers.
Fiscal policy decentralization is of particular relevance, since it establishes fundamentally new principles for the formation and use of state financial resources at the level of territorial-administrative units. An important component of fiscal policy is the assessment of its level of decentralization. Researching the process of implementing the fiscal policy of the state in the context of the reform of decentralization of powers, it should be borne in mind that it is first necessary to assess the current state of financial independence of local budgets in three

LITERATURE REVIEW
The basic theoretical and methodological aspects of decentralization in general and fiscal policy decentralization in particular were explored by many economists. Thus, Oates (2005), one of the founders of the theory of fiscal federalism, author of the so-called decentralization theorem, proposed a "second-generation fiscal federalism theory" in which he substantiated the changes in the state activity of governmental fiscal institutions in the context of federalism. Dabla-Norris (2006) suggested the basic principles of fiscal policy decentralization based on an analysis of the implementation of decentralization in countries with economies in transition, and also identified the risks of the negative impact of decentralization on the effectiveness of the national economy. Freitag and Vatter (2008) examined the negative impact of fiscal policy decentralization on the budget arrears of Swiss territorial units, arguing that such an impact occurs only during a recession. Kyriacou, Muinelo-Gallo, and Roca-Sagalés (2013) examined how the quality of government power mediates the relationship between fiscal decentralization and regional imbalances. Volokhova (2014) summarized the theoretical foundations of the local finances decentralization, improved the method of assessing the fiscal decentralization state and financial support of own and fixed revenues, intergovernmental transfers of local budgets. Sydor (2014) identified and substantiated possible directions for strengthening the financial autonomy of local budgets in order to effectively execute local and national delegations of powers by local governments. Baskaran, Feld, and Schnellenbach (2016) analyzed the impact of fiscal federalism and decentralization on economic growth. Savy, Pauliat, and Senimon (2016) examined the decentralization processes in Europe, concluding that they are non-linear and different, depending on the size of the region, the traditions of self-government, the role of the central government, and other political, social, economic, and cultural characteristics. Bartolini, Stossberg, and Blöchliger (2016) have argued that fiscal policy decentralization reforms must be two-sided: the growth of local budget revenues should be matched by adjustment of intergovernmental transfers and fiscal equalization. Bartolini, Stossberg, and Blöchliger (2016) examined the relationship between regional inequality in the countries and the state of fiscal decentralization. The results of an analysis of 30 OECD countries in 1995-2011 have shown that in those countries where local government budgets are mainly financed by local taxes, local resources are used more effectively, and fiscal policy is stepping up territorial development. Krysovatyi and Desiatniuk (2016) formed the directions of tax policy of Ukraine in the context of fiscal decentralization and expansion of the tax base. Alonso and Andrews (2018) have argued that fiscal policy is more effectively implemented under decentralized conditions and found a negative relationship between socio-economic deprivation and efficiency. Bellofatto and Besfamille (2018) examined the optimal fiscal decentralization level in the countries with a federal system, according to the analysis of the administrative capacity and fiscal capacity of the regions. Melnyk, Sineviciene, Lyulyov, Pimonenko, and Dehtyarova (2018) examined how fiscal decentralization affects the stability of the economic system. Kim and Dougherty (2018) explore how intergovernmental budgetary relationships affect the economic growth and income distribution in the Netherlands, South Korea, India, and the United Kingdom. Vozniak (2019) explored the peculiarities of financial decentralization reform in the context of its impact on the sustainable growth of regions, etc.
Separate aspects of decentralization of power are devoted to the works of Shevchuk (2013), Martynenko (2015). However, the problem of integral assessment of the level of fiscal policy decentralization is insufficiently addressed and needs thorough research. So, the main purpose of the article is to improve the methodological support for assessing the level of fiscal policy decentralization based on an integrated approach.

DATA AND METHODS
In order to assess the level of fiscal policy decentralization in the dynamics, it is advisable to use a methodological apparatus of integrated assessment, which will allow us not only to comprehensively assess the fiscal policy decentralization, but also to carry out a comparative analysis of its components with the further development of measures that will solve the identified problems. The main source of estimates is the public statistics on official statistics posted on the official websites of CASE Ukraine (2019) and State Statistics Service of Ukraine (2019). In the course of the analysis, the necessary calculations will be made of the necessary single indicators, sub-indices and comprehensive integral indicator of diagnostics of the level of decentralization of fiscal policy of Ukraine. It is also advisable to define the critical boundaries and interpret the high, medium, low, and critical levels of fiscal policy decentralization.
To implement a methodological approach to the integrated assessment of the level of fiscal policy decentralization, first, the indicators will be identified and grouped (Table 1).
An important methodological apparatus for assessing the effectiveness of fiscal policy in the context of fiscal policy decentralization is an integrat-

Name of indicator
Method of calculation Nature of influence*

Revenue sharing ratio (R RS )
The ratio of local budget revenues to consolidated budget revenues  Note: * stimulant is an indicator whose effectiveness in increasing dynamics; destimulator is an indicator whose performance is in decreasing dynamics. ed approach. The integrated approach involves the development of an integral indicator, which is the weighted average of sub-indices formed based on the functional principle of fiscal policy implementation in the context of fiscal decentralization.

Sub-index of the level of decentralization of intergovernmental budget relations (Î DIBR )
We propose to calculate the integral indicator (I) by the formula of geometric mean: Thus, the formula of the sub-index, which includes single indicators of fiscal policy effectiveness and level of fiscal decentralization, is as follows: Standardization of indicators is carried out, having previously determined which of them are stimulants (the indicator increase influences the improvement of the state of entrepreneurship) and destimulators (positive is the indicator's decrease). The formula used to standardize the stimulant indicator (Z ij↑ ), is as follows: The formula for standardizing the destimulator indicator (Z ij↓ ) is as follows: The standardization process, which is implemented according to formulae (4) and (5), ensures that all indicators are brought to one dimension in the range from 0 (the worst value) to 1 (the best value).
The methodological support of the weighting coefficients calculation is the most difficult stage of the integral assessment. If to consider the logic of calculating the weight of single indicators, the weight of an individual indicator is the degree of its influence on the integral index, which consists of several more indicators, so it can be argued that the most affecting integral indicator is the component that has the greatest influence on the rest of the components (Omelyanenko, Martynenko, Slatvinskyi, Povorozniuk, Biloshkurska, & Biloshkurskyi, 2019).
The impact of one indicator on another is measured by the value of the paired correlation coefficient, which expresses the degree of closeness (density) of the correlation relationship between a pair of indicators and can acquire the values of r ∈ [-1; 1]. In this case, the value r = -1 reflects the functional feedback, an increase of the value of one indicator leads to a corresponding decrease of the value of the second indicator; a value of r = 1 indicates a direct functional relationship (an increase of the value of one indicator leads to a corresponding increase of the value of the second indicator); r = 0 means that there is no complete relationship between the indicators as such. For other values of the paired correlation coefficient other than ±1 and 0, the value of the correlation relationship is determined -it will be larger when approaching ±1.
However, to determine the weight of each indicator introduced into the integral indicator, the direction of their action (inverse or direct effect) can be neglected. To do this, the modules of values of paired correlation coefficients will be taken. Paired correlation coefficients can be obtained by constructing a correlation matrix using statistical data processing software. To determine the degree of interaction of a particular indicator with other indicators, it is necessary to summarize the modules of all correlation coefficients associated with this indicator. By the sum of the modules of the pair correlation coefficients, it is possible to rank the indicators from the most to the least significant. This is also confirmed by research- And so on.

RESULTS
Using the data in Table 1, the single indicators will be calculated that will be introduced into the integral indicator of the level of fiscal policy decentralization within their groups (Tables 2-4).  In 2017, the ratio of provision own income to GDP reached its maximum, reaching 7.7% of GDP at actual prices. Cost coverage ratio (ratio of own revenues of local budgets (without transfers) to expenditures of local budgets) and ratio of financial autonomy (ratio of own revenues of local budgets (without transfers) to revenues of local budgets (including transfers)) reached maximum values in 2004, and minimum -in 2014, but during 2015-2017, they showed positive dynamics. Table 3 summarizes the indicators reflecting the state of fiscal policy decentralization by the structure of local budgets. Thus, during 2004-2017, the tax and non-tax revenue ratios, which interpret the share of tax and non-tax revenues in total lo-  cal budget revenues, showed negative dynamics -2.2% and -1.4%, respectively. Also, the rate of official transfers, which shows their share in local budget revenues, increased by 9.4%, which is a negative trend due to the modernization of fiscal policy in the context of fiscal decentralization. However, the ratio of overall tax stability and revenue base stability ratio in 2017 reached a historical high and had a positive effect on fiscal policy decentralization. Table 4 summarizes the indicators reflecting the impact of intergovernmental transfers on the fiscal decentralization of local governments. Thus, in the dynamics of 14 years, there was a reduction of the financial independence ratio (the ratio of own revenues of local budgets (without transfers) to intergovernmental transfers) by 0.38 points. The coverage ratio of expenditure transfers increased by 0.15 points, the budget sustainability ratio increased by 0.38 points, which is a negative trend and reflects the increased role of intergovernmental transfers in financing the revenue side of local budgets. This is also pointed out by a 0.2 percentage point decrease in the stability ratio of budget revenues section. However, the local budget deficit ratio during 2015-2017 was a positive figure, reaching a maximum of 0.07 in 2017, which makes it possible to conclude on the overall local budget surplus.
Thus, to assess the level of fiscal policy decentralization in Ukraine, its main components were formalized, which revealed the main tendencies in three key areas that contributed to or hindered the modernization of fiscal policy in Ukraine. It should be noted that unambiguous conclusions can be drawn about the level of fiscal policy decentralization, and, therefore, there is a need for an integrated assessment.
The integral assessment of the level of fiscal policy decentralization starts by constructing correlation matrices for each component of fiscal policy decentralization in order to calculate the weighting coefficients according to formulae (6) and (7). One can build a correlation matrix using the MS Excel "Data Analysis," the function "Correlation." As a result, the impact of fiscal decentralization on the formation of local budget revenues is shown in Table 5. From the data given in Table 5, it can be seen that the closest correlation between five indicators of the fiscal policy decentralization by the influence on the formation of local budget revenues between the ratio of provision own income and ratio of provision own income to GDP (r = 0.641), the ratio of provision own income and the ratio of financial autonomy (r = 0.686), the cost coverage ratio and the ratio of financial autonomy (r = 0.892). In this case, the smallest correlation is between the revenue sharing ratio and the ratio of provision own income (r = 0.112).
A correlation matrix of 6 individual indicators of the level of component of fiscal policy decentralization was constructed according to the structure of local budgets (Table 6). As can be seen from Table 6, the closest is the direct correlation between the tax revenue ratio and the ratio of overall tax stability (r = 0.803), the inverse correlation coefficient between the tax revenue ratio and the rate of official transfers (r = -0.831) and also the direct correlation of the ratio of tax autonomy and the stability revenue base ratio (r = -0.921). The least dense is the inverse correla-tion between the tax revenue ratio and the stability revenue base ratio (r = -0.081).
A matrix of partial coefficients of paired correlation of five single indicators of decentralization of intergovernmental relations is constructed (Table 7). Among the partial coefficients of paired correlation given in Table 7, the highest level of inverse correlation density (r = -0.987) is available between the financial independence ratio and the budget sustainability ratio. The lowest level of inverse stochastic coupling (r = -0.232) is presented between the stability ratio of budget revenues and the local budget deficit ratio.
The calculation of the weighting coefficients of individual indicators of the level of the fiscal policy decentralization by influence on local budget revenues formation is presented in Table 8.
As can be seen from the data in Table 8, the most significant single indicator in the component of fiscal policy decentralization by influence on local budget revenues formation was the fourth one -the cost coverage ratio (w 4 = 0.247), and the least significant one -the revenue sharing ratio (w 1 = 0.155). The sum of the weighting coefficients of five single indicators of the component of the level of fiscal policy decentralization by the influence on the formation of local budget revenues was one. Therefore, the calculations made are correct and will be used in the formation of the sub-index of the level of fiscal policy decentralization by the influence on the formation of local budget revenues.
Similarly, Table 9 and Table 10 are built to calculate the weighting coefficients of single indicators of the component of the level of fiscal policy decentralization for the structure of local budgets (Table 9) and the weighting coefficients of single indicators the component of the level of decentralization of intergovernmental budget relations (Table 10).
As can be seen from the data in Table 9, the most significant single indicator in the component of fiscal policy decentralization for the structure of local budgets was the third one -the rate of official transfers (w 3 = 0.201), and the least significant one was the non-tax revenue ratio (w 2 = 0.119). The sum of the weighting coefficients of six individual indicators of the component of fiscal policy decentralization for the structure of local budgets was one. Therefore, the calculations made are correct and will be used in the formation of the fiscal policy decentralization sub-index by structure of local budgets.  As can be seen from the data in Table 10 Table 11.

Total
As can be seen from the data in Table 11   It should be noted that standardization of indicators is carried out in order to reduce their numerical values to one unit of measure, which greatly facilitates the development of an appropriate integral indicator.

Total
Similar calculations are carried out for the other component of the level of fiscal policy decentralization (Tables 12 and 13). Table 14 is the result of a methodological approach to the integrated assessment of the level of fiscal decentralization. Table 14 summarizes the sub-index equations of the components of the fiscal decentralization level and the integrated integral indicator. It should be noted that an integral indicator of the level of fiscal decentralization is proposed as the geometric mean of three sub-indexes (see Formula (1). In the authors' opinion, the main advantage of the statistical tool "geometric mean" is the limits within which the integral indicator can be: I LFPD ∈ [1 ;0].  The size of the interval is defined to break the levels of fiscal policy decentralization into 4 groups, using the following formula:

The result of calculations in
where i -the size of the interval, X max -the highest value of the indicator, X min -the lowest value of the indicator, n -number of groups.
After calculations, we get the following levels of fiscal decentralization: • 0 ≤ I LFPD < 0.25 -the critical level of fiscal decentralization, at which the degree of obstacles to the development of the fiscal policy of the state is maximum; • 0.25 ≤ I LFPD < 0.5 -low level of fiscal decentralization, at which the degree of obstacles to the development of fiscal policy of the state is significant; • 0.5 ≤ I LFPD < 0.75 -the average level of fiscal decentralization, at which the degree of obstacles to the development of fiscal policy of the state is acceptable; • 0.75 ≤ I LFPD ≤ 1.0 -high level of fiscal decentralization, at which the degree of obstacles to the development of fiscal policy of the state is minimal.
Thus, for economic interpretation of the numerical value of both the sub-index of the appropriate  component and the integral indicator of the level of financial decentralization, the division at levels -from critical to high -is presented. It is clear that, at a critical level of fiscal policy decentralization, obstacles to fiscal policy modernization will be maximal and, at a high level, obstacles will be kept to a minimum.

DISCUSSION
According to the calculations, using the equations given in Table 14, the integral indicator of the level of fiscal decentralization is calculated, and its economic interpretation and validation in the dynamics of the last 14 years are carried out (Table  15). Table 15 shows that during 2004-2017, the minimum level of fiscal policy decentralization was fixed in 2014 (І LFPD = 0.143), including the sub-index of the level of fiscal policy decentralization by impact on local budget revenue generation was critically low (Î DLBR = 0.196), as well as the sub-index of the level of decentralization of fiscal policy by the structure of local budgets (Î DSLB = 0.155) and the sub-index of the level of decentralization of intergovernmental relations (Î IBRD = 0.097). The maximum level of fiscal policy decentralization was in 2007 (І LFPD = 0.71), which is at the average level, with sub-indices of the level of fiscal policy decentralization in terms of influencing local budget revenues and local budgets on intergovernmental fiscal transfers reached a high level. That is, it can be concluded that the level of decentralization of fiscal policy in Ukraine during the studied period fluctuated from the critical to the average (Figure 1). Figure 1 shows that the largest steady decline in the integral indicator of the level of fiscal policy decentralization in Ukraine occurred during 2004-2012 -from the average to the critical one. It is also clear from the graph that the decision to implement the decentralization reform was delayed, but forced, as the level of fiscal policy decentralization reached a minimum in 2014. As a result, over the next three years, the level of decentralization of fiscal policy in Ukraine has been raised to the average (І LFPD = 0.58). Therefore, in pursuing the modernization of fiscal policy in Ukraine, it is necessary to find the levers to minimize obstacles and risks that hinder the fiscal policy decentralization. Given the results obtained, it should be noted that the problem of introducing an integrated approach to assessing the level of fiscal policy decentralization, the solution of which this article is devoted to, is new and is being implemented by the authors for the first time. The proposed integral indicator of the level of fiscal policy decentralization is a universal tool that can be used in many ways. In particular, it can be used to rank the regions of the same country or to rank the different countries in terms of fiscal policy decentralization.

CONCLUSION
Thus, in implementing the methodological approach to assessing the level of fiscal policy decentralization in Ukraine, an integral indicator of the level of fiscal policy decentralization was constructed as a geometric mean of three sub-indices: the level of fiscal policy decentralization by influence on the formation of local budget revenues, the level of fiscal policy decentralization on the structure of local budgets, the level of decentralization of intergovernmental relations. As a result, at the end of 2017, compared to 2004, the level of fiscal policy decentralization in Ukraine declined but remained at an average level for which the risks of fiscal policy modernization were acceptable, compared with the lowest numerical value in 2014, which corresponded to critical level of the integral indicator with significant obstacles to the fiscal policy modernization.
Thus, the level of fiscal policy decentralization in Ukraine decreased by five times during 2004-2014, including the level of fiscal policy decentralization by the influence on the formation of local budget revenues decreased by three times, the level of fiscal policy decentralization by structure of local budgets decreased by four times, the level of intergovernmental budget relations decentralization decreased by 5 times. The level of fiscal policy decentralization in Ukraine increased by four times during 2014-2017, including the level of fiscal policy decentralization by the influence on the formation of local budget revenues increased by three and a half times, the level of fiscal policy decentralization by structure of The proposed and tested methodological approach should be at the forefront of government regulation and modernization of fiscal policy in Ukraine. The results obtained during its development lay the foundations for finding the effective ways of enhancing the financial independence of local authorities, forecasting the trends, and developing the scenarios for the further development of fiscal policy, as well as ensuring the effectiveness of its modernization in the uncertain environment.