Pages:     | 1 |   ...   | 8 | 9 || 11 | 12 |   ...   | 16 |

- transfer growth by one unit results in expenditure increase by 0,5 up to 1,5, which depends upon the year (besides, statistically significant dependence can be observed only for the variables within one and the same year).The fact that during some years expenditures increase (or decrease) to a greater degree than financial aid amount does can be explained by several reasons. Firstly, it is possible that expenditure matching on account of tax revenue growth be undertaken (within the years when expenditures growth is larger that transfer increase tax revenue also grows). Secondly, expenditure can be covered by non-tax revenues, which are also changing year after year. Thirdly, regarding the equation like (61) and (62) we can assume that other factors determinant for expenditures and tax revenues do not change year after year but this assumption may fail to be true. This drawback can be offset by the right specification of the model.

In addition to regression analysis qualitative analysis of changes in transfers impact on regional revenues and expenditures was carried out (see Appendix). This analysis showed that for most of the regions transfer modification results in similar to it in sign expenditure modification, but for a much smaller number of the regions it leads to the opposite tax revenue modification.

As it was demonstrated above, the analysis of lump-sum general grant impact on the choice of the authorities, which turn to be grant-recipients, between public and private goods proves the leakage effect of some resources available attained by the increase in regional private goods consumption (tax rate decrease). Alongside with that theoretical analysis of the model shaped proves that financial aid growth always results in tax revenue decrease and expenditure growth, when α value is less than β (financial aid allocation is based upon actual tax revenues rather than actual expenditure value). Besides, if β parameter sufficiently exceeds α along with financial aid growth expenditures decrease attained by substitution effect proves to be possible.

The results of the calculations done in order to define the impact of the transfer growth on expenditures show that theoretical analysis coincide with empirical data. But the analysis of the effect produced by equalizing transfers upon regional tax revenues in the Russian Federation fails to prove any statistically significant dependence.

In order to explain absence of federal financial aid leakage in the private sector in Russia it is possible that the assumptions made about some other empirically given phenomenon such as flypaper effect interpreted in the literature on public finance and fiscal federalism be applied. This effect can be characterized by the phrase Money sticks where it hits.36 According to this effect the response of lump-sum grant recipient observed while his choosing between public and private goods proves to be different from the response of the voters to income growth on the same territory (e.g. as a result of the decrease in federal taxes paid in this region). This difference is experienced even though both situations (the grant received by the regional authorities and private agents’ income growth) prove to be equivalent from the point of view of theoretic analysis. Thus, the investigations of general grant impact (here: block grants) on local public expenditures conducted for the USA show that 100$ grant received by the local government leads to expenditure growth by 40-50$ caused by public goods consumption as long as equivalent private agents’ income growth attained by federal tax burden relief on the territory results in local expenditures growth only by 5-10$37.(37)

In order to explain the situation, when lump-sum grant leads to a larger expenditure growth than it can be expected in theory, it is possible that Niskaken’s model for bureaucrats’ behavior be made use of38. According to this model regional bureaucrats maximize the budget they bear responsibility for as long as it is assumed that their reward, administration power and other benefits are determined by the budget size. Such premise shows that optimum for the bureaucrat budget size (based on average benefits being equal to average costs) will exceed efficiency level achieved by the equality between marginal benefits and marginal costs. If it is assumed that the region proves to be lump-sum grant recipient, then it is in regional bureaucrats’ interest to persuade the political authorities of acting as if it were a categorical specific grant. It results in the authorities’ failure to provide tax reduction along with the increase in private goods consumption39(39), which could be expected if the choice were made according to the representative voter’s interests.

Then, the deviation of the results gained by the grant from theoretical assumptions can be explained by fiscal illusion effect40. If it is assumed that the population of the region choose the level of public goods production on the basis of tax price subjective assessment (the correlation between tax liabilities and public goods provsion supply in the region), then lump-sum grant can be considered by the regional authorities as general increase in regional income. In this case public goods provision level is increasing (by the value dependent upon the elasticity of the demand for public goods with respect to income) as long as the level of taxation is decreasing due to the maintaining of the same tax price (at the same average costs within public goods provision). But it is also possible that regional authorities offer public goods to the population at a new subsidized tax price, which causes public goods consumption increase based upon the elasticity of demand for public goods (corresponding to the population needs under the condition of the information being available) according to the price.

Thirdly, it is possible that the effect of the increase in public goods consumption attained by lump-sum grant being more than it could be expected be determined by institutional reasons such as implicit dependence of the grant as well as high transaction costs induced by changes in tax rate (especially if it is expected that revenues growth is temporary and taxes should be increased in the future anyway)41.

The reasons mentioned above could fail to be sufficient to explain an extremely high expenditure increase attained by transfer to the regions in Russia (average annual expenditures growth is close to 100% of the total sum of federal grants received). Such a profound influence exerted by financial aid upon regional budget expenditures may be caused by some additional reasons. First of all, it should be noted that budgetary system in Russia be characterized by a huge amount of budget debts outstanding to the suppliers of goods and services as well as to social payments recipients. Thus, actual demand for public goods in the regions achieves the level exceeding actual expenditures for public goods provision, which is especially characteristic of the regions greatly in need. Under such conditions the lump-sum grant proves to be specific (or even matching) for current legislation obligates the regional authorities to provide public goods and grant social transfers to the needy greatly exceeding regional own revenues and financial aid received.

It should also be noted that actual conditions of federal financial support be a bit different from the given model. For instance, while calculating and allocating transfers, the federal authority requires from the regional administration to maximize regional tax rate, to collect tax arrears, to transit budget execution into the federal treasury, etc. Such restrictions provide against grant leakage into the private sector.

It can be also assumed that regional budget expenditures for public goods provision as well as social situation in the region be more important for regional authorities than benefits gained by tax burden relief regarding insignificant tax incidence variation between the regions.

From the analysis of the model of regional authorities’ behavior it can be concluded that as long as the weight assigned to actual expenditures less important than 'normative' expenditures ( α value decrease) tax revenues should also decrease as well as it can be concluded that the sign of change in expenditures is not defined and depends upon the relationship between the parameters. If fiscal capacity is more important for financial aid allocation procedure than actual tax revenues ( β value decrease), tax revenues and actual expenditures prove to be increased.

In order to analyze the impact of..a..and..b.. modification on revenues and expenditure value selected by the regions it is necessary that modification signs of..a.and..b be compared with modification signs of revenues and expenditure within adjacent years. To draw the comparison the regions increasing or decreasing their revenues in 1995-98 should be calculated. The comparison made will regard revenues and expenditures per capita in comparable prices (according to minimal subsistence level). The results of the calculation (in % to the whole number of the regions) are given in the table below:

Tax revenues












Share of regions with observed increase











Share of regions with observed decrease











Judging by the results of the estimation done for equation (56) we can arrive at the conclusion that in 1994-98 all the three parameters α, β, γ prove to be falling. Alongside with that there was no common trend in tax revenues and expenditures for partial derivatives of optimal tax and expenditure values with respect to these parameters have different signs and are not defined. For instance, the indefinite sign of tax revenue and expenditure modification is determined by the fact that within simultaneous modification of α and β parameters (as it was shown in the theoretical section) relative dynamics of the latter prove to be more important than their absolute value. And, besides transfer value, revenues and expenditures also depend upon some other factors (expenditure policy parameters, taxation reforms, etc.)42 Another way to define the influence exerted by the parameters upon tax revenue and expenditure is to perform econometric estimations with respect to panel-data, in which α, β, γ parameters will function as variables explaining either constant or changeable values within a year (e.g. regression (56) residuals given in terms of different α, β, γ. for different regions.)

* * *

Thus, analyzing the results received through empirical verification of the hypotheses about financial aid allocation principles generated in 1994-1999 and the impact of the latter on regional fiscal behavior we can arrive at the following conclusions.

1. Within the given time period the allocation of financial aid from the federal budget was aimed at covering some particular part of the gap range between regional budget revenues and expenditures. The empirical data do not contradict the hypothesis that corresponding gap estimations were done by fiscal authorities on the basis of both actual regional revenues and expenditures and regional fiscal capacity and expenditure needs. As time passed, financial aid allocation procedure grew more dependent upon fiscal capacity and expenditure needs rather than upon actual revenues and expenditures. The econometric estimates, which regard temporal lags, prove that 'normative' value impact on fiscal gap calculations is getting less considerable as long as the lag between the calculation and financial aid allocation is becoming smaller, which corresponds to actual characteristics of budget procedures.

2. The estimations prove the hypothesis that while allocating financial aid the federal center is orientated to actual tax revenues than actual expenditures. In other way it can be formulated like orientation to expenditure needs rather than to regional budget fiscal capacity. While interpreting such an assumption it is necessary to consider that official transfer allocation methodology up to 1999 was based on the one hand upon actual tax revenues and on the other upon adjusted expenditure values for 1991. Besides, low tax revenues (with low taxing powers of the regionalgovernments) are more determinant for increase in transfer amount than high regional budget expenditures.

Pages:     | 1 |   ...   | 8 | 9 || 11 | 12 |   ...   | 16 |

2011 www.dissers.ru -

, .
, , , , 1-2 .