FiSCAL MANAGEMENT ANd PoLiTiCAL BUdGET CYCLES iN TRANSiTioN ECoNoMiES: dATA PANEL ANALYSiS

In this paper, on the basis of relevant statistical tests, the influence of the electoral process on the trajectory of fiscal indicators in the transition countries is analyzed. The aim of the research is to identify the political manipulation of certain fiscal policy mechanisms in transition countries.The focus of the survey is on the growth of general government spending, the reduction of general government revenues and the creation of budget deficits as the coherent consequences of fiscal expansion in the pre-election period. By testing, there is no relevant evidence of the use of tax incentives as a form of political action on the economic sphere. On the other hand, the results of the survey indicate that in the observed countries, there really is a rise in government spending in the period before the election process and, consequently, the growth of budget deficits. However, according to the same findings, in the post-election period there is no reduction in consumption. Growth in general governments consumption continues, but to a lesser degree, which in turn leads to the correction of the budgetary balance.

makers remain at the disposal of fiscal policy in an attempt to artificially improve the economic results in the pre-election period or, more often, in the case of misuse of fiscal policy, reallocate income and improve the position of certain segments of society in accordance with their party-ideological or opportunist motives (Jakšić & Praščević, 2011). The current government seeks to improve the chances of re-election in the future by stimulating aggregate demand in (current) pre-election times.
In this regard, during the election years, political representatives are implementing an expansive fiscal policy, in the form of reduction of tax liabilities, increase of social and other transfers, and directing state spending towards projects with high direct visibility of results. The results of the use of fiscal mechanisms for political purposes are manifested by the growth of the budget deficit, the increase of public spending and the change in its composition and the reduction of public revenues. Periodic fluctuations in the fiscal structure are called political budget cycles (PBCs) and their occurrence is related to the cyclical nature of electoral processes (Shi & Svensson, 2003, 2006.
The subject of research in the work is PBCs, presented as a result of political abuse of fiscal measures and instruments in transition countries. It is emphasized that the electoral system of the given country is a key generator of PBC.
The aim of the research is to analyze the amplitude of the basic fiscal indicators in the form of public spending, public revenues and budgetary equilibrium, to determine if and to what extent there is a political influence on the conduct of fiscal policy in the mentioned countries, which is primarily stimulated by the election cycles.
The work paper is divided into five sections. After the Introduction, which outlined initial concerns of the research problem, the second part focuses on the literature review. In the third part, the methodological basis of PBC testing was presented. In the fourth part, the results of the research and the estimated hypothesis were presented. In the fifth, concluding part, the results, arising from the research process, were summed up.

THE REViEW oF LiTERATURE
The use of fiscal policy for political purposes is most often proven by identifying the growth of public spending, reducing tax burdens, or changes in the structure of budgetary expenditures during the preelection period. Until recently, the empirical evidence of the election abuse of fiscal policy did not highlight significant PBC distinctions among countries, while recent approaches highlight the groundbreaking differences in the way, intensity, frequency and duration of the use of fiscal mechanisms for political purposes. The modality of PBC, according to De Haan and , can be created by factors such as the level of economic development, the quality of institutions, the age and level of democratic institutions in the country, electoral rules and forms of governance, the transparency of the political process, the check and balances regulation, i.e. a system of mutual balance and budget control and fiscal rules.
Findings on the patterns of PBC expression in the context of the level of development of countries and the character of their political frameworks make a particularly illustrative perspective on the modes of manipulative use of fiscal policy.
The evident manifestation of PBC in the form of increased total consumption is verified in so called new economies (Brender & Drazen, 2005) and developing countries (Shi & Svensson, 2006). The reasons for this, on the one hand, can be found in the underdevelopment of the political system of these countries where voters do not know enough about the functioning of the political process, and, on the other hand, in motives that drive politicians in the conduct of fiscal policy. It is indicative that in the initial stage of the development of the democratic system, voters are not fully able to correctly anticipate the effects of pre-election fiscal policy in order to make the right choice. Such circumstances lead politicians to an attempt to deceive in the way of using fiscal policy, without expecting that they will have significant consequences for this, given that they will secure a position in power by winning elections, and that in the postelection period they will probably have a restrictive fiscal policy.
Although, generally speaking, the artificial adjustment of economic results is characterized as an inherent phenomenon for developing countries (Streb, Lema & Torrens, 2005), it is possible, based on the attitude of the dominant influence of the electoral process on the implementation of fiscal policy, to confirm the presence of PBC within a heterogeneous group of countries, from poorly developed democracies, to OECD countries .
The weaker presence of PBC in developed countries is explained through several basic assumptions. In stable democratic systems, pre-election fiscal indiscipline most often encounters a negative reaction among voters, whereby the growth of the budget deficit in the election years actually reduces the likelihood of re-selection of the ruling nomenclature (Brender & Drazen, 2008;. A significant number of studies (Brender, 2003;Drazen & Eslava, 2010;Garmann, 2017) denote fiscal conservatism as a limiting factor in the re-election of political structures that are populistically shaped in the way of implementing economic policy. The size of the cycle is reduced under the influence of the development of democratic conditions, greater transparency and symmetric distribution of information (Akhmedov & Zhuravskaya, 2004). The volume of fiscal manipulation depends on the asymmetry of information and ignorance of voters in (pre-electoral) circumstances, given that in conditions of incomplete and inadequate information, politicians try to exploit the lack of information and uncertainty among voters. By contrast, informed voters, aware of the possibilities of appropriating political rents, tend to limit ineffective forms of behavior by political representatives. In addition, the system of fiscal rules, as a characteristic form of regulatory co-ordination of fiscal policy in developed countries, largely inhibits political misuse of budgetary resources (Rose, 2006). Also, PBCs are a less arbitrary outcome of the abuse of fiscal policy in economies that are more oriented to foreign trade, since in such circumstances management is more demanding (Schuknech, 1996).
PBCs, in the context of the time maturity of the pre-election period and economic expansion, can be intensified by the behavior of the electorate. Namely, in conditions of strong economic growth, voters may demand an increase in public spending and a reduction in taxes, thereby encouraging the pro-cyclical nature of fiscal policy towards the current economic trajectory (Alesina, Campante & Tabellini, 2008).
Although some information indicates that in the previous period governments of the EU countries implemented stabilization policies (Andrikopoulos, Loizides & Prodromidis, 2004), recent findings warn of the growing use of fiscal policy for electoral purposes. Listening to ideology as an important factor in making political decisions in electoral years, Doležalová (2013) seeks for elementary mechanisms of action of political actors in the process of winning the electorate's inclination. According to the research on the EU23 sample it can be concluded that current incomes constitute a key means of fiscal manipulation of ideologically mixed governments, and that left wing politicians are predominantly focused on the expenditure side of the budget. Similarly, the tendency towards undertaking the preelection fiscal initiative is largely determined by the level of fragmentation of political institutions.
Undoubtedly, a high concentration of political power is a context in which there is no effective system of restrictions and penalties for the use of current economic mechanisms for political purposes. In support of such a statement, Boyar (2015) offers evidence suggesting that in the system of the rule of one or more groups of political parties, there is a ramification of the abuse of fiscal policy.
In a sense, the misuse of fiscal policy is spurred by the unique monetary framework of the EMU. According to Donahue and Warin (2007), the channeling of political intentions through the monetary sphere is limited in the candidate countries, or actually disabled in EMU member countries. In accordance with such limitations, an alternative form of political maneuvering of economic policy is achieved through the fiscal system. Certain number of studies really confirms the presence of PBC within the EMU (Mink & De Haan, 2006). Besides this, the extent to which politicians are willing to generate PBCs, violating the given fiscal norms, depends on a compromise between the benefits of re-election and the costs of fiscal irresponsibility (Efthyvoulou, 2012). The results show that a good part of the functioning of the EMU is reflected in the process of convergence of national budget deficits and the reduction of discretionary fiscal policy management (Fatás & Mihov, 2003), and that direct election manipulation of fiscal policy more often occurs in conditions of favorable macroeconomic environment (Golinelli & Momigliano, 2006).
It is indisputable that the scientific contribution of this paper is accelerated by a relatively small number of studies on manipulation of fiscal policy in post-socialist systems (Klašnja, 2008;Mačkić, 2014), but also by efforts to valorize the effects of their political transformation in the process of transitional evolution and detect the real impact of political mechanisms on economic performance.
In accordance with the subject and defined goals of the research, the following research hypotheses are tested: H 1 : Increase in total public spending is the result of pre-election manipulation of fiscal policy in observed transitional economies.
H 2 : The decrease in total public revenues is the result of pre-election manipulation of fiscal policy in transition economies.
H 3 : The growth of the budget deficit is the result of pre-election manipulation with fiscal policy in transition economies.
In relation to the basic hypotheses and assumptions about oscillations of fiscal indicators that are conditioned by the election cycles, conducted research hypotheses are tested: H (1.1) : In the post-election period, there is a fiscal contraction, or a decrease in total public spending in the observed transition economies.
H (2.1) : In the post-election period, there is a fiscal contraction, or an increase in total public revenues in the observed transition economies.
H(3.1): In the post-election period, there is a fiscal contraction, i.e. a reduction in the budget deficit (surplus growth) in the observed transition economies.
According to the set of research hypotheses, and in accordance with the economic theory of PBC, Table 1 presents the election and post-election effects on fiscal policy.
During the election cycle public expenditure is raised, but public income is lower, so the result is deficit in balance. Conversely, after the election cycle, public expenditures are declining, revenues are rising, so there is a surplus in the budget.

AN oVERViEW oF FiSCAL PoLiCiES iN TRANSiTioN ECoNoMiES
In order to further analyze, it is useful to observe the relative performance of the analyzed countries in terms of their performance from the aspect of tax policy and public spending policy. Dispersion diagrams ( Figure 1 and Figure 2) point to the degree of efficiency of fiscal policies in the domain of tax regulation (stimulation) and the expansive public spending policy in the observed transition economies.
The tax policy index consists of three subindicators of tax regulations: marginal personal income tax, marginal corporate income tax and total tax burden in relation to GDP. The value of the tax burden index is calculated by means of the quadratic cost function, in order to present the declining yields in the case of very high taxation rates. The value obtained on the 0-100 scale is transmitted using the following formula: The tax burden index ij = 100 -α (Factor ij ), where i is the country observed, j represents tax rates (marginal rates for individual and tax incomes and total tax burden in relation to GDP), and α represents a coefficient whose value is constant.
According to the basic rules of statistical conclusions, a greater distance from the coordinate start should indicate that a country with a successful tax policy (greater distance from the x axis) achieves a high level of GDP per capita (greater distance from y axis). The presented data point to the different effects of the tax system, i.e. the different levels of GDP per capita as a form of final economic effect. Countries such as Georgia, Armenia, Kyrgyzstan, Moldova, Tajikistan, Ukraine and Uzbekistan, despite the highly assessed ways of managing tax policy, do not achieve significant levels of living standards. Better results, regarding the establishment of an efficient tax system, are 209 Lj. Radonjić / SJM 15 (2) (2020) (2) (2020)   The Public Expenditure Index refers to consumption and transfer payments for different programs, without indicating the "optimal" level of public expenditure. However, certain guidelines exist, with reference to recommendations on the adequate use of fiscal policies that will neither affect inefficiently economic growth, nor lead to dislocation of resources. The presented data also indicate a significant heterogeneity among the countries observed. It is evident that countries like Kyrgyzstan, Moldova, Tajikistan, Ukraine and Uzbekistan have the worst performances in fiscal policy use. Similarly, as in the tax policy analysis, the method of implementing fiscal policy is somewhat better assessed in the case of the Western Balkans, while the Baltic and Central European countries achieve the best results in the management of fiscal channels.

ANALYTiCAL FRAMEWoRK ANd METHodoLoGY oF RESEARCH
The study empirically tests the political generation of PBCs. The central part of the analysis relies on estimating the movement of fiscal variables within and outside the election process. The survey is conducted on the basis of an unbalanced panel consisting of 28

Fiscal sector
The paper starts from the assumption that PBCs represent cyclical trajectories of public spending and public revenues as the outcome of the use of fiscal measures for political purposes. Fiscal policy in pre-election conditions is characterized by a significant growth of so-called discretionary expenditures. These expenditures provide "competence" to economic policy makers who deceive voters (Praščević, 2013). On the other hand, political abuse can take the form of reducing tax burdens, which as a rule results in a reduction in budget revenues. The categories of fiscal variables of total public spending, total public revenues and budget balance are indicated by the parameters EXPENDITURE, REVENUE and BUDGET, respectively. The values show the level of consumption, income and their equilibrium as a percentage of gross domestic products (GDP). The data were collected from the World Economic Outlook International Monetary Fund's statistical database (2019).

Election process
For the purpose of registering electoral periods, dummy variables are introduced. ELECT variable represent the election year, while POST-ELECT represents a period in which the election process is not held, including every year, except for the electoral one. The values of the artificial variables are binary, i.e. ELECT = 1 for the year considered electoral and ELECT = 0, for all other years in which the electoral process is not conducted. The POST-ELECT = 1 variable is valid for the year after the election, while POST-ELECT = 0 has a given value for all other periods in the observed interval. It should be kept in mind that, in the conditions of realization of the electoral process at the beginning, that is, in the first quarter of the election year, potential pre-election manipulation occurs in the previous calendar year. In this regard, in the work for electoral processes conducted in the first three months of the current year, the election year is considered the previous year.

Control variables
Control variables are used in order to control more adequately a certain number of relevant economic indicators that can affect the fiscal system. The basic control parameters are the growth rate (GROWTH), real gross domestic product (GDP) and gross domestic product per capita (GDPpc) expressed in international dollars according to purchasing power parity (PPP). We use all three indicators in order to minimize the impact of periodic fluctuations in economic activity on the results of the research. For easier manipulation in the testing process, the values of the GDP variables are expressed by natural logarithm (lnGDP). GDP per capita is the GDPpc parameter. In order to eliminate fluctuations in the fiscal policy induced by the effects of business cycles, the trend of the cyclical GDP growth of each country (TREND) is analyzed. The trend of the business cycle is extracted from data on annual GDP values, using the Hodrick-Prescott filter. The HP filtering parameter is 6.25. The data on the current account balances of the observed economies are expressed as a percentage share in GDP and are determined by the BALANCE parameter. Data for GDP, GDP per capita and current account balance were derived from the World Economic Outlook (IMF) International Monetary Fund statistical database (2019). Also, the analysis examines the impact of the shocks arising from the process of international trade of countries, and for this purpose, an indicator on the level of the percentage share of trade in the GDP of the country (TRADE) is introduced. Trade is the sum of exports and imports of goods and services measured as a share of gross domestic product. Data for economic growth and trade were taken from the World Bank's World Bank's World Development Indicators (2019) statistical base. The impact of the demographic factor (demographic aging) on fiscal policy is perceived by using the age dependency coefficient (RATIO). This indicator measures the ratio of economically dependent persons (under 15 and over 64) and the working age population (aged 15 to 64). Data on age dependency coefficients were derived from the World Bank's Statistical Health Nutrition and Population Statistics (2019). Control variables are incorporated into each type of analysis.

Model
Correlation of the movement of fiscal aggregate indicators and election cycles is examined using an adequate econometric model (Figure 1). Persson and Tabellini (2003) offer several concepts that can be useful in evaluating parameters. The test is performed on the basis of a general regression equation of the type: (1) where: Y i,t -is fiscal (dependent) variable, i.e. state consumption and total revenues, X -is the vector of control variables dummy -is artificial variable related to electoral process, electoral pattern, political and party system. λ, β -are regression coefficients ɛ i,t -is a standard error The regression equation is tested by using the static program Stata 12. The model estimation is done through a fixed effect model.

RESULTS oF RESEARCH ANd diSCUSSioN
The analysis conducted on the sample of 28 countries in the period 1995-2017 covers a total of 146 electoral processes. The structure of unbalanced data panels is shown in Table 2.
The central part of the analysis certainly refers to the testing of the PBC model, which is done by the equation (1). Table 3 presents the descriptive statistics of the analyzed variables. Table 4 shows the parameters of regression testing, defining the influence of pre-election and post-election periods on fiscal variables.
In economic theory, it is emphasized that economic policy makers, since they know the preferences of voters, carry out fiscal expansion and/or avoid the necessary increase in tax liabilities, which will be well accepted by voters. Consequently, in the post-election period there is a rise in tax burdens to finance pre-election expansion, which will not be foreseen even by rational entities because of the complexity of the budgetary process, on the one hand, and the limitations of information on the other. The outcome of this kind of action by politicians is PBCs. conclusions. In the pre-election period and during the electoral process, public spending is rising (0.65), while in post-crisis conditions, instead of reduction, it continues to register slightly slower growth (0.62). Similarly, in pre-election circumstances, public revenues are growing (0.25), which is intensified in the post-election period (0.7). It is illustrated that the most important change is recorded within the budget balance, bearing in mind that during the pre-election period a budget deficit is achieved, while the surplus in the budget is achieved in the post-crisis stabilization environment in the observed countries.
According to regression outputs, it is possible to evaluate research hypotheses. Considering the slight increase in total consumption in the pre-election period of the H 1 hypothesis (Increase in total public spending is the result of pre-election manipulation of fiscal policy in observed 214 Lj. Radonjić / SJM 15 (2) (2020)    However, as in post-election circumstances there is no decrease in consumption, but it only slowly grows, hypothesis H 1.1 (In the post-election period, there is a fiscal contraction, or a decrease in total public spending in the observed transition economies) was rejected. On the other hand, the model does not confirm the theoretical assumption of a reduction in public revenues before the election. As the hypothesis H 2 (Reducing total public revenues is the result of pre-election manipulation of fiscal policy in transition economies) is not valid, it is automatically rejected and the hypothesis H 2.1 is made (In the post-election period, there is a fiscal contraction, that is, an increase in total public revenues in the observed transitional economies). Although the result suggests that public revenue rises in the post-election period, they did not fall (according to the theoretical assumption) in the electoral cycle, which automatically excludes the possibility of accepting the hypothesis H 2.1 . In the end, the results indicate that the growth of the budget deficit is conditioned by the way of fiscal policy in the pre-election period, and that the postelection fiscal restriction leads to budget surpluses. In this sense, the H 3 hypothesis is accepted (Budget deficit growth is the result of pre-election manipulation of fiscal policy in transition economies). Diterioration of fiscal discipline caused an increase in public debt and deepining in budget deficits of countries (Hilal & Yildirim, 2014). Also, the H 3.1 hypotheses is acepted (In the postelection period, there is a fiscal contraction, i.e. a reduction of the budget deficit (surplus growth) in the observed transitional economies). As in the post-election period there is a decrease in public expenditures and an increase in public revenues, as a final effect, there is a surplus in the budget.  Of course, with the assumption of manipulative use of fiscal measures in preelection circumstances, the effects of the global economic crisis on the way of implementing fiscal policy should also be taken into account. The slowdown in the global economy, the problems of the financial sector and the functioning of the financial system, which significantly hampered global lending and reduced liquidity, also had negative effects on emerging markets and economies in transition. These consequences were related to the problem of reducing global aggregate demand, which resulted in a slowdown in economic activity and a decrease in exports, as well as a decline in capital inflows in the form of investments and a rise in borrowing costs due to a reduced global liquidity. In such circumstances, the implementation of anti-recession measures was largely based on fiscal policy mechanisms. With the return of the country's active role in the economic sphere through fiscal stimulation, political actors have again become responsible for the economic policy and the results of that policy.
For transitional economies, it is important to point out that the emergence of political cycles is predominantly caused by political institutional solutions and the impacts of political institutions on the economic sphere. Numerous studies confirm, on the one hand, a significant positive link between economic growth and key political institutional conditions such as the protection of property rights, political stability, state efficiency and the rule of law, and, on the other hand, the negative impacts of institutional weaknesses on the long-term economic dynamics (Piatek, 2016). The significance of institutional harmonization is reflected in the experience of transition countries and the need to implement an optimal reform strategy, which recognizes the interdependence of the economic and political system.
There is a two-way determination between the political cycles and the given institutional framework. Political cycles can have strong effects on economic performance which either directly, or through the challenge and maintenance of instability, makes the establishment of efficient institutional solutions more difficult or impossible. Continuous manipulation with economic policy instruments affects the depth of the crisis, generating high economic and social costs, especially evident in the process of institutional transformation.

CoNCLUSioN
The paper examines the existence of political budget cycles in 28 transition countries. The analysis was carried out on the case of an unbalanced panel of samples, during the period 1995-2017. The aim of the study was to examine the role of the political system in determining the cyclical movement of key fiscal aggregates through an appropriate econometric model. In this respect, an empirical analysis of the behaviour of total state spending, total state revenues and budget balance at time intervals before and after the election process was conducted.
The results of the analysis are far from simple and one-sided conclusions. Regression parameters on the one hand prove that there is a (slight) increase in total public spending in pre-election conditions, but also that, on the other hand, in the postelection period there is no restriction of spending, that is, public expenditures continue to grow at a lower rate. It is indicative that the model does not confirm the theoretical assumption that one of the types of pre-election manipulation of fiscal policy concerns the reduction of tax burdens, that is, the reduction of total public revenues. In the pre-election and post-election period, public revenues on the sample of analysed economies are growing. Finally, the hypothesis that the budget deficit is growing in pre-election, while decreasing in the preelection context is fully confirmed.
It should be noted that political cycles are expressed in short-term time intervals, starting from the pre-election period, to postelection implications of political abuse of economic policy. Considered from the aspect of the influence on key macroeconomic values, political activity in the electoral context is most often manifested in shortterm (monthly, semi-annual, annual) and very rarely in long-term fluctuations.
PBC research in transition countries is a very important segment of recognizing political and economic interaction in contemporary social systems. In circumstances that involve the construction and consolidation of democratic political institutions on the one hand and the institute of a modern open economy on the other hand, the possibility of political abuses in the process of implementing fiscal policy is growing. In conditions of insufficient control of political representatives and institutional non-construction, manipulation with fiscal policy is certain.
The emergence of political cycles is also induced by the high degree of constrained rationality of voters in transition countries. When voters are not able to adequately assess the effects of certain policies, political structures allow for greater room for manipulative activities. Certain transitional systems are characterized by a certain degree of political instability, which in turn leads to frequent electoral processes, that is, preconditions for political abuse of economic policy. It is indisputable that fiscal policy mechanisms should be kept within the framework of actions aimed at achieving valid economic results and to limit the political influence and use of fiscal instruments for political purposes as much as possible. Fiscal policy driven by political motives most often leads to overindebtedness of the public sector during the election period, which ultimately leads to macroeconomic anomalies that manifests itself in the problems of functioning of public services, a decline in investment generally slowing down economic activity in the post-election period. Also, discretionary fiscal policy can lead to distrust of economic actors in the institutions of the system, which can contribute to low investment activity, relocation of resources and weak economic growth in the future. Therefore, keeping fiscal policy independent of political interests is an important element in the functioning of the national economy and the generation of positive macroeconomic results.