REGIONAL PROTECTION IN RUSSIA: HARMFUL OR BENEFICIAL?

Interim report

Evgenia Kolomak

Statement of purpose

Regional leaders were inclined to interference in the regulation of regional economy and to protection of local producers almost everywhere in Russia. Among the political tools, which the regional policy-makers have at their disposal, the intensity of local market price control and subsidies to producers differs essentially.

The project has an aim to study how significant the barriers erected by the regional administrations for interregional flow of goods are. The project also considers the consequences of the regional protectionism for economic development of the territories.

Review of literature

I. Theoretical issues

Literature on the economics of transition regards the price liberalization as a key element of transition, because it is a necessary condition for the introduction of the market mechanism and for the improvement in the allocation of resources. These studies deal with important efficiency dimension to price liberalization (Lipton and Sachs (1990), Boyko (1992), McKinnon (1991)). However, there is also an important redistributive dimension to price liberalization, and this consideration is a political economics’ argument (Drazen (2000), Persson and Tabellini (2002)). What legislators should put in place depends on the political acceptability of the reforms. Roland (2000) writes that the political constraints affect the speed and design of reforms.

One of the aspects of the redistribution that price reforms cause is between regions. The between regions distributive dimension to price reform is a topic of political economy of international trade policy (Krugman and Obstfeld (1994), Caves, Frankel and Ronald (1996)) where one of the discussions is focused on national welfare arguments for and against protectionism. Applying the results to transitional economy Kruegel and Ciolko (1998) demonstrate that the hypothesis of the endogeneity of price liberalization variable can not be rejected. The worse the initial conditions for transformation, the greater the probability of the deep transformation recession is, and hence there are more likely delays in liberalization. When initial conditions are favorable, rapid liberalization is feasible and preferable. Castanheira and Popov (1999) also suggest that the speed and extend of price liberalization may be endogenous, liberalization policy may depend on the initial conditions and magnitude of the decline in output as a result of liberalization.

The political constraints are reinforced when the fact that bureaucrats and regulators may benefit from the persistence of price control is taken into account. Shliefer and Vishny (1992) applying a rent-seeking model, show that price control creates shortage rents for state sector and represents opportunities for soliciting bribes. Another explanation of low prices is given by Berkowitz (1996). The theory developed in this paper assumes that local politicians are motivated to serve constituents rather than to collect bribes. The model predicts that local government would choose a market clearing price when, firstly, private sector is monopolized, and, secondly, the share of nonresidential consumption within a jurisdiction is high.

So the political constraints may make a gradual price reform preferable despite its efficiency costs (Dewatripont and Roland (1992 a, b) Roland (2000)). Milder reforms are the only way to speed up the process and enhance political acceptability. Bertocchi G. and M. Spagat (1997) give another explanation for gradualism and reversion in the price reforms, applying learning model they show that instability, which economies in transition are faced, dampens learning effect and make policy more gradual.

The economic distortions resulting from partial price liberalization are discussed in Murphy, Shleifer and Vishny (1992). The partial price reforms encourage diversion of inputs away from the regulated sector towards enterprises that are less constrained by arbitrarily regulated prices. When only a few regions impose maximum prices, they hurt themselves as well as the producers, and benefit enormously the regions that have not restrained prices which simply free ride on them.

Summarizing literature on price liberalization Castanheira and Popov (1999) write that there is no theoretical evidence that a big bang approach is associated with less cost than the gradual one and there is no persuasive empirical evidence that fast liberalization performed better than slow one. Liberalization should and would lead to better performance in the longer run.

Features of local policy depend also on the attitudes towards the governments. Paper by Edwards and Keen (1996) synthesizes the two extremes: the view of government as a Leviathan and the view of government as a benevolent maximiser of their citizens’ welfare. The policy-makers have quasi-concave preferences defined over some item of public expenditures which, while financed from general revenues, benefits only the policy-maker, and the welfare of their representative citizen. Polishchuk (2000) shows that under certain assumptions a revenue-maximizing Leviathan-type government might offer better conditions for economic growth than a benevolent, which is concerned about economic well-being of its constituency at large.

The role of lobby groups in the shape of trade policy is incorporated into analysis in two ways. The first approach stresses political competition between opposing candidates. In the works of Magee et al. (1989), Hillman and Ursprung (1988), the lobby groups evaluate their prospects under the alternative trade policy proposals have been made by competing parties. In making their giving decisions, the lobbies weigh the benefit of an increasing the probability of their favorite party being elected against the direct cost for the donation. The parties use the resources to influence the election outcome. In the second approach presented in Stigler (1971), Hillman (1982), Grossman and Helpman (1994), the economic policy is considered as being set by an incumbent government seeking to maximize its political support. The political support function has as arguments the welfare that designated interest groups derive from the chosen policies and the deadweight loss that the policies impose on society at large. In this formulation, campaign contributions do not enter directly into the analysis, and the political competition of the next election is kept in the background. Both of the approaches consider the political optimization as underlying the endogenous determination of trade policy.

II. Stylized facts and estimations for Russia

1. Tools of the regional regulation includes as direct (price ceiling, making-up price, limitations on profitability, limitations on trade extra charge, declaration about change in price) as indirect methods (tax exemptions, credits, subsidies, budget compensations) and differ essentially among the regions. (Appendixes 2, 3).

The TACIS report ranks regions by price control in October 1995. The survey is conducted in capital cites. The ranking differentiates between types of price controls: from subsidies and limitations on the profitability and price mark-ups to rationing and issuance of coupons. The ranking is based on seventy-three food goods surveyed by Goskomstat. Tallying up points for different categories of price controls, regions are rated from most liberal (1) to most regulated (1/47).

2. Growth of private wholesale trade and «porosity» of the interregional borders have undermined attempts of some regional administrations to implement an autarchy policy in food supplying. For instance, in 1995-1996 the famous “Ylyanovskaya” system of control over price and supply of food failed and was discontinued (Õåíñîí (2001)). However the sub-federal documents erect barriers and limitations on interregional flow of goods and services by introduction of recurring certification, marking and “identification”, by direct prohibition of export from and import in the region (Âåðõîâåíñòâî Êîíñòèòóöèè ÐÔ è … (2000).

3. Documents adopted at sub-federal level in Russia regulate prices and tariffs on a wide range of regional state and private production (Appendix 1).

4. Price differentiation among the regions is very high in Russia. In 1998 variation of regional living earnings was 30,9%, in 1997 the variation of indexes of consumer prices was 10%, indexes of electric power price – 18%, values of consumer basket – 47% and its share in population’s money income – 50%. The dynamics of these indicators are presented in table 9 Appendix 3.

5. There are not only high thresholds for interregional trade, which are surprising for neighbouring regions, because transport costs do not explain them. There are also observed diverged time series of price levels in the regions. (Ãëóùåíêî (2001))

6. A “culprit” of the weak market integrity of Russia is a behavior of a group of regional economies, which lag behind in price liberalization and preserve subsidizing to domestic producers. The economic indicators in these regions are worse in comparison with similar jurisdictions. (Áåðêîâèö, Äåéîíã (2000), the analysis is based on the approach by Engel and Rogers (1996)).

7. Almost all regional policy-makers in Russia base micro-management at their jurisdictions (not always unselfish) on protection of domestic enterprises and on resistance to expansion of the external ones (Õåíñîí (2001)). As federal budget subsidies decreased as regional authorities join more actively in supporting enterprises (table 1, 2, Appendix 3). However, the level of subsidizing differs essentially among the regions (table 3, Appendix 3).

8. The main source of hidden subsidies to enterprises is as follows: power subsidies in the form of overdue payments and of barter with artificially high prices; non-payments in budget and non-budget funds, state purchase paid by barter or by tax exemptions (Äîêëàä Âñåìèðíîãî áàíêà (2000), Êîëîìàê (2001), tables 1, 2, 5, Appendix 3).

9. An important feature of the debts and barter is that they provide the ground for the regional and local executive authorities to control the enterprises and level of their profitability. In exchange for the cooperation the policy-makers soften budget constrains of the enterprises and protect against external competition. More over barter became profitable business for mediators, including bureaucrats of the executive bodies. It resulted in a wide spread of scheme for extracting of benefits, of collusion and corruption. The extraction of political rent became a reason for the highest regional decision-makers to protect the enterprises (Äîêëàä Âñåìèðíîãî áàíêà (2000)).

10. However, the existence of tight local budgets does not explain why many regions allocate scarce financial resources to subsidize basic consumption goods (Berkowitz (1996)). Inter-budgetary relation is a factor provoking barter and subsidies. The level of transfers from upper level budget to lower level budget depends on actual expenditures of previous year. This makes regional bureaucrats interested in exaggeration of expenses, in order to have a possibility to obtain more transfers next year, or to provide a higher share of taxes distributed between different budgetary levels. Barter payments is an easy way of reaching this aim. At the federal level such practice was stopped in 1999-year budget (Äîêëàä Âñåìèðíîãî áàíêà (2000)).

The empirical estimations have shown positive and statistically significant correlation between subsidies granted to industrial, agricultural and constructing enterprises and VAT transfers (table 7, Appendix 3), between tax exemptions to branches of industry and transfers (table 8, Appendix 3) and between subsidies and level of price regulation (table 6, Appendix 3).

11. A characteristic of the regional budgets is high level of overdue for salary and transfers to population (more 40%), the next item is overdue to infrastructure monopolies, supplying public utilities (28%). Hence, the biggest part of burden, resulted from hidden subsidies to enterprises is imposed on population, what contradicts to thesis that subsidizing and help to enterprises are explained by social imperatives of the regional authorities (Äîêëàä Âñåìèðíîãî áàíêà (2000)).

Methodology

I. A Model

Statement of the problem

We consider a regional market, so we may assume that the economy is small and market regulation is the result of the political process. One of the attitudes of Russian regional economies is a high level of specialization in the production, the producers have incentives to form lobby groups and they demonstrate ability to overcome the free-rider problem.

The regional lobby groups confront regional policy-makers with requirements to provide protection for the sector against external producers in exchange for political support. The regional government bears costs for implementing an inefficient protection policy that is result of creating deadweight loss and its accountability to the general electorate. The government sets protection policy comparing benefits of the political cooperation with local producers and costs of deterioration of its reelection prospects. The implemented policy must be financially feasible.

The proposed theoretical framework for the analysis of the barriers of regional price regulation is very similar to the one developed by Grossman and Helpman (1994) in the study devoted to protection trade policy.

Overview of Grossman - Helpman’s results

Grossman and Helpman consider a small, competitive economy. Free trade is efficient for such an economy, so any policy interventions can be ascribed to the political process. They assume that there is a high degree of concentration in the ownership of the specific inputs and that the various owners of some these inputs have banded together to form lobby groups. They assume also that some factor owners overcome the free-rider problem to conduct joint lobbying activity, while other do not.

The lobby groups may offer political contributions to the incumbent politicians, who are in a position to set the current trade policy. While the lobby groups ignore the effects of their contributions on the election probabilities, the incumbent politicians may see a relationship between total collections and their reelection prospects. Incumbent politicians’ objective is to maximize a weighted sum of total political contributions and aggregate social welfare.

The authors model the lobbing process as follows. Each interest group confronts the government with a contribution schedule. The schedule maps every policy vector that the government might choose (where policies are import and export taxes and subsidies) into a campaign contribution level. The government then sets a policy vector and collects the contribution associated with its choice.

Let introduce some notations: p is the vector of domestic prices; Ci(p) - the contribution schedule tendered by lobby i; Wi(p) - gross-of-contributions joint welfare of the members of lobby group i; G(p) - government’s utility function.

The authors are interested in the political equilibrium of a two-stage non-cooperative game in which the lobbies simultaneously choose their political contribution schedules in the first stage and the government sets policy in the second. An equilibrium is a set of contribution functions {Ci0(p)}, one for each organized lobby group, such that each one maximizes the joint welfare of the group’s members given the schedules set by the other groups and the anticipated political optimization by the government; and a domestic price vector p0 that maximizes the government’s objective taking the contribution schedules as given. The Nash-equilibrium contribution schedules implement an equilibrium policy choice.

Grossman - Helpman’s model has the structure of a menu-auction problem. Bernheim and Whinston (1986) have characterized the equilibrium for a class of such problems. Grossman and Helpman applied these results to the problem of protection trade policy. The adaptation resulted in following proposition.

Proposition 1. ({Ci0}iÎL, p0}) is a subgame-perfect Nash equilibrium of the trade policy game if and only if:

(i) Ci0 is feasible for all iÎL;

(ii) p0 maximizes G(p) on the set of domestic price vector;

(iii) p0 maximizes Wj(p) - Cj0(p)+ G(p) on the set of domestic price vector for every jÎL;

(iv) for every jÎL there exists a pj that maximizes G(p) on the set of domestic price vector such that Cj0(pj)=0.

Condition (i) states that lobby’s contributions must be nonnegative and no greater than the joint income available to the sector. Condition (ii) states that, given the political contributions offered by the lobbies, the government sets trade policy to maximize its own welfare. Condition (iii) stipulates that for every lobby, the equilibrium price vector must maximize the joint welfare of that lobby and the government, given the contributions offered by other lobbies. Condition (iv) requires that for every lobby j there must exist a policy that elicits a contribution of zero from lobby j, which the government finds equally attractive as the equilibrium policy p0. If there does not exist such a policy, then lobby j can lower their political contributions without changing the government’s choice, what of necessity leave sector j strictly better off.

Condition (iii) characterizes the equilibrium structure of protection. Condition (iv) characterizes the equilibrium structure of political contributions.

Our problem and one of Grossman - Helpman are very similar and we largely rely on the significant results obtained by the authors, however there are several differences. The differences come from three issues. The first one is the fact that Russian regional governments can not use export and import tariffs and subsidies opposed to the case of Grossman - Helpman consideration and are restricted to other tools of price regulation: price ceiling, price mark-ups, price subsidies, tax exemptions or credits. The second issue stems from the requirement of financial acceptability of the regional protection policy, regional budget constrain needs explicit introduction into the model. The third difference is explained by the statement of problem to distinguish between different tools of the protection policy. These differences modify Grossman - Helpman‘s model and obviously its analytical inferences as well.

Formal framework

We consider a regional market with tradable goods i=0,1,…,n. The local demand curve for a particular good is di(pi). Assume when there is no price dispersion all consumers prefer domestic goods. If the regional government does not interferes into the market-clearing mechanism the regional market faces exogenously given prices p0*, p1*,…, pi*,…, pn*. Suppose that in the absence of trade the equilibrium price of goods i=1,…,n is higher than in the situation of interregional and/or international trade.

The supply curves of local producers depend on input and output exogenous prices and/or implemented local protection policy. Assume the regional government can use input and output subsidies, input and output price limitations, and tax exemptions. We assume that production in each sector requires labor and a specific input, subsidized are and regulated are prices of the specific inputs. Consequently the supply function of a locally produced good i depends on price (which differs form the exogenous market price if local government imposes price limitations), input subsidy, output subsidy, input price restrictions, and tax exemption yi(pi, si,ri,ti) where pi is target price, si is subsidy per unit of good i, ri is subsidy per unit of specific input in sector i, and ti - tax exemptions granted to sector i. Let denote by p, s, r,, t  the vectors of output prices, output and input subsidies, and tax exemptions respectively.

Let the regional economy is populated by individuals with identical preferences. Each individual maximizes utility given by

                                                                                                                                       (1)

where x0 is consumption of good 0 and xi is consumption of good i, i=1,…,n. Good 0 is a competitive and does not need a protection, let its price equals to 1.With these quasi-linear preferences an individual spending an amount E consumes xi=di(pi) of good i, i=1,…,n, where the demand function is inverse of ui¢(xi), and . The consumer surplus derived from the goods is equal to

                                                                                                                     (2)

Where p is the vector of the prices targeted on the protection of local producers.

The protection must be financially acceptable. The financial acceptability means satisfying the budget constrain, local government expenditures should be less than receipts. The receipts are in the form of taxation of the domestic aggregate income, the expenditure items are input and output subsidies to local producers. The excess of receipt over protection expenditures, here regarded as a source of public expenditures financing, is:

                                                                                    (3)

Where t is the tax rate, zi(pi, si, ri, ti) - demand for specific input in sector i, and b reflects the local government ability to ‘soften’ local budget, it can be done through transferring expenses of local policy to another budgets or by obtaining additional resources from higher level budgets.

The producers are interested in protection for their sectors and enter the political activity. The lobby representing a sector i makes its political contribution contingent on the protection policy implemented by the government. Denote by ci(pi,si,ri,ti) the contribution tendered by lobby i. The lobby determines the contributions to maximize total welfare of the sector’s members: labor income plus profit of the sector plus consumer surplus and benefits from the public expenditures less contributions. The scheme of the distribution of the political donations among the sector’s members is out of consideration here. We assume the existence of ways to allow all the members to share the gains from the political coordination. The joint gross-of-contribution welfare of the members of sector i is:

                                                                  (4)

Where pl - the wage rate; pi* - exogenous price of the specific input in sector i; ai - share of the voting population related to sector i.

The government’s utility function depends on attitudes towards government. There are two extreme types of government presented in the literature as stark alternatives: benevolent and Leviathan. When the government is a benevolent, it is a maximizer of their citizens’ welfare. A Leviathan – government maximizes items of expenditures benefit only the policy-makers. A more general assumption is that policy-makers are neither wholly benevolent nor wholly self-serving, an obvious encompassing is that the policy-makers maximize a weighted sum of citizens’ welfare and their own wellbeing. The latter assumption is adopted to our problem.

The incumbent government maximizes a weighted sum of political contributions and aggregate welfare of the population. The political contributions provide direct benefits to the government. However the social welfare can result in indirect benefits if voters are more likely to reelect a government that provides a high standard of living. The government objective function is:

                                                                                   (5)

Where 0£q£1.

We consider a two-stage non-cooperative game: in the first stage sector’s lobbies make decision and propose political contributions contingent on protection policy; in the second stage the regional government determine the implemented policy. An equilibrium is a set of contribution functions {ci0(pi,si,ri,ti)}, one for each sector, such that each one maximizes the joint welfare of the sector’s members given the schedules proposed by the other sectors and the anticipated optimization by the regional government; and a regional protection policy vector (pi0,si0,ri0,ti0) that maximizes the government’s objective taking the contribution schedules as given. The Nash-equilibrium realizes an equilibrium policy.

The proposed formal framework corresponds to the structure of Grossman-Helpman’s problem. However the modification of the Grossman-Helpman’s model and more detailed consideration of some issues modify Proposition 1. The proposition relevant to our problem is as follows.

Proposition 2. ({ci0}iÎL, {p0,s0,r0,t0}) is a subgame-perfect Nash equilibrium of the regional protection policy game if and only if:

(a) {ci0} is feasible for all iÎL;

(b) {p0,s0,r0,t0} maximizes G(p,s,r,t) subject to budget constrain r(s,r)³0 for all iÎL;

(c) {p0,s0,r0,t0} maximizes Wj(p,s,r,t)+ G(p,s,r,t) subject to budget constrain r(s,r)³0 for every jÎL;

(d) for every jÎL there exists a bundle (pj,sj,rj,tj) that maximizes G(p,s,r,t) subject to budget constrain  such that cj0(pj,sj,rj,tj)=0.

The equilibrium structure of protection policy

Grossman and Helpman have proved that if the contribution schedules are differentiable around the equilibrium, the shape of the political contributions reveal the lobbies’ true preferences in the neighborhood of the equilibrium. They have also demonstrated an interesting property of Nash equilibria, in equilibrium government behaves as if it attributed to lobbies higher weigh than other population. Below we show that these results hold to our model as well.

Let assume that the political contribution functions and welfare functions are differentiable. To characterize the structure of the equilibrium protection policy let consider conditions (b) and (c) Proposition 2, they imply that the first order condition is satisfied at {p0,s0,r0,t0}:

                                                                        (6)

                                                                                                (7)

Where l is a Lagrange multiplier. Inserting (7) into (6) gives . By definition (4) . Taken together the equations imply

                                                                                                     (8)

Equation (8) establishes that around the equilibrium change in the political contributions reflects the effect of change of the government protection policy on the joint welfare of members of the lobby’s group.

By the definition (4) , where Wi(p0,s0,r0,t0) is net-of-contribution welfare of group i members. If the political contributions correspond to true preferences of the group, than Wi(p0,s0,r0,t0)³Wi((p,s,r,t) and

                                                                                                          (9)

Condition (b) of Proposition 2 states that if (p0,s0,r0,t0) and (p,s,r,t) are feasible than , or . From expression (9) .

Consequently the government in the equilibrium maximizes weighted sum of welfare of different groups of population. Welfare of groups of population presented by lobbies in the political process receives weight (1+q), welfare of other ones receives weigh q, where 0£q£1.

Further let present in a more detailed record expression (7), it takes form . Inserting (8) into the expression gives

                                                              (10)

The equation shows how marginal change of protection policy influences the welfare of the groups of populations distinguishing between participating in lobbing and do not participating.

So the properties of the equilibrium structure of the regional protection policy are as follows. Firstly, around equilibrium the political contributions reveal preferences of the interest groups regarding protection policy. Secondly, equilibrium protection policy results in distribution of welfare in favor of the sectors, participating into political lobbing. Thirdly, in equilibrium marginal change of welfare of different groups influenced by the protection policy depends on the fact of participating in political lobbing.

At the current stage of the analysis we consider the cases when the regional government is restricted to one of the protection tools: output subsidies, input subsidies, target prices, or tax exemptions.

a) Output subsidies

Let first consider output subsidies, we analyze a solution of equation (10) when the region faces exogenously given prices pi=pi*, input subsidies equal zero ri=0, tax exemptions equal zero  ti=0, and output subsidies only are at the disposal of the regional government si³ 0.

From (4) we find  , where sij - Kronecker’s symbol. Substitution of the terms in expressions (10) allows to derive  Let introduce an indicator variable ji that equals 1 if the sector uses lobby pressure and 0 - otherwise. Denote by L. The equation takes the form .

From (3) we find . Inserting let to derive , where ejs - subsidy elasticity of production good j.

Proposition 3. The government in the equilibrium chooses output subsidies that satisfy

   for all j=1,…,n

So output subsidies for a good positively correlated with tax rate, subsidy elasticity of production, ability of the regional administration to soften regional budget constrain, with weight attributed to population’s welfare, exogenous input price and with lobbing activity of the sector. And output subsidy for a particular good negatively correlated with level of exogenous output price.

b) Input subsidies

This case supposes to analyze the equation (10) when pi=pi*, si=0, ti=0, and ri³ 0.

From equation (4) , equation (10) takes form:

Or .

From (3) we can find . Taken together let us derive , where ejr - input price subsidy elasticity of demand for input of production good j.

Proposition 4. The government in the equilibrium chooses input subsidies that satisfy

      for all j=1,…,n

So input subsidies for a good positively correlated with level of the exogenous input price, with tax rate, with input price subsidy elasticity of demand for input, ability of the regional administration to soften regional budget constrains and lobbing activity. However level of input subsidy for a particular good negatively correlated with output exogenous price.

c) Tax exemptions

This case supposes analysis of equation (10) when pi=pi*, si=0, ri=0, and ti³0.

From (4) we find . Substitution of the terms in expressions (10) let us to derive The equation takes the form .

From (3) we find . Inserting lets to derive , where ejt - tax exemption elasticity of production good j.

Proposition 5. The government in the equilibrium chooses tax exemptions that satisfy

  for all j=1,…,n

So tax exemptions for a sector positively correlated with tax rate, tax exemption elasticity of the production, with weight attributed to population’s welfare, with lobbing activity of the sector and with input exogenous price. Tax exemption for a sector negatively correlated with level of exogenous output price.

d) Target price

This case supposes analysis of equation (10) when si=0, ri=0, ti=0, and pi is fixed at a target level.

From (4) we find . Substitution of the terms in expressions (10) let us to derive

The equation takes the form .

From (3) we find . Taken into account that local demand is satisfied with local production and import . So . From (2) we derive

Inserting gives the result , where ejp - price elasticity of production good j.

Proposition 6. The government in the equilibrium chooses prices that satisfy

 for all j=1,…,n

So target prices for a sector positively correlated with level of import and negatively correlated with level of exogenous input price, price elasticity of output, and marginal utility of good.

The table below summarizes the characteristics of the equilibrium protection policy when one instrument of protection .is used.

Table 1. The correlation characteristics of equilibrium

Variables

Output subsidy

Input subsidy

Tax exemption

Target price

Tax rate

+

+

+

 

Exogenous output price

-

-

-

 

Exogenous input price

+

+

+

-

Weight attributed to population’s welfare

+

 

+

 

Overall lobbing pressure on a regional government

+

+

 

 

Political activity of the sector’s lobby

+

+

+

 

Output subsidy elasticity of production

+

 

 

 

Input price subsidy elasticity of demand for input

 

+

 

 

Tax exemption elasticity of production

 

 

+

 

Price elasticity of output

 

 

 

-

Level of import

 

 

 

+

Marginal utility of production

 

 

 

-

Ability of the regional administration to soften regional budget constrain

+

+

 

 

II. Empirical estimations

A. Hypotheses

Assuming the model is correct the empirical estimations will support the hypotheses as follows.

Hypothesis 1. Regional price regulation is a feature of regions that have weak competitive positions and higher level of import.

For testing these hypotheses connection between price regulation and characteristics of efficiency of regional economic development will be estimated.

Hypothesis 2. Subsidies, tax exemptions and price regulation depend on exogenous input and output prices, so macroeconomic demand and supply shocks resulted in raise of prices may also result in change of protection activity.

Data on regional subsidies and price regulation confirms this hypothesis (Appendix 3, table 2). There were growth of price regulation activity in 1995 and 1998 and increase of share of total subsidies in 1998 These years are famous for sharp devaluation of ruble, growth of consumer demand and prices of goods of both import and domestic production.

Hypothesis 3. Regional subsidizing and granting tax exemptions provoke higher tax burden in the regions.

For testing these hypotheses connection between subsidies, tax exemptions and level of the regional tax collection will be estimated.

Hypothesis 4. Regions demonstrating active subsidizing of output have larger share of transfers from federal center and higher level of non-payments in their budgets.

To test the hypothesis the dependence of level of transfers received by region and of budget non-payments on price regulation will be estimated. The preliminary estimations have shown positive significant correlation between subsidies and transfers from the federal budget (Appendix 3, table 7).

Hypothesis 5. Lobbying power depends on the concentration of the producer’s interests, the higher is the concentration the higher is the ability to influence the government and to persuade it of the protection.

To test this hypothesis the correlation of the regional subsidizing and tax exemptions and level of the regional specialization will be estimated.

Hypothesis 6. Regional price regulation depends on import pressure in the region.

To test this hypothesis the correlation of the regional price regulation and level of the regional import will be estimated.

B. Methods of estimation

The theoretical model structure implies to make an empirical analysis by sectors and by regions. Because of radical changes during transition period in Russia time effect must be taken into consideration as well.

The updating of the empirical information is still in the process, and at the present stage of the research the preliminary empirical estimations have been done for the aggregated by sectors data. The observations have the panel structure and include characteristics of the regions (about 80) over time period (1996 – 2000), let i=1,…,N is index for region, and t=1,…T is index for time. The system of the hypotheses shapes the system of regression equations.

        (A)

    (B)

(C)

C. Information

Testing of the formulated hypotheses assumes data on subsidies, tax exemptions, price regulation by regions and by sectors, and information on structure of economies, efficiency of production, on budgets, taxes, and import by regions.

Detailed structure of regional budget incomes and expenditures for 1996 - 2000 are taken from Ministry of Finance of the Russian Federation. The characteristics of regional economic development, including dynamics, price level, efficiency and structure of the production are presented in statistical yearbook "Regions of Russia ". Information on tax exemptions and price regulation is reflected in database “Consultant +. Regional legislation”.

D. Results of the estimations

For the empirical estimations were generated variables as follows:

-         share of subsidies for industry and agriculture in regional budget expenditures in region i in year t;

-         number of decisions aimed at direct price regulation in region i in year t;

-         number of decisions granted different tax exemptions in region i in year t;

-         dummy variable for the macroeconomic shock, which takes the value “1” for year 1998;

-         dummy variable for the specialization level, which takes value “1” when there is an industry producing more than 1/3 of total regional industrial product in region i in year t and value “0” other wise;

-         share of transfers from federal budget in the regional budget income in region i in year t;

-         tax income per capita in total regional budget income in region i in year t;

-         share of unprofitable firms in industry in region i in year t.

The results of the empirical estimations are presented in the tables below.

Table 2. The results of regression (A)  estimation

Variables

Within estimation

GLS estimation

coefficient

P-value

Coefficient

P-value

Constant

7.117

0.063

5.926

0.13

Dummy for macroeconomic shock

-4.742

0.150

-5.381

0.023

Dummy for specialization level

0.322

0.836

1.388

0.025

Share of transfers

0.057

0.512

0.116

0.001

Tax income per capita

0.001

0.488

0.001

0.004

Share of unprofitable firms

-0.102

0.138

-0.110

0.020

R2

0.204

0.311

Breusch-Pagan test for random effects

c2(1)=8.67, P-value=0.0032

Hausman specification test

c2(5)=2.07, P-value=0.8395

Hausman test shows that we can rely on GLS estimations that are more effective. So the correlation of the share of subsidies for industry and agriculture in regional budget expenditures subsidies with the generated variables is statistically significant.

The obtained estimations have confirmed that the high level of taxation, lobbing activity, and transfers form higher level budget are the factors allowing the regional governments to increase the subsidizing of the local firms. The calculations confirmed the significance of the macroeconomic shocks for the subsidizing and revealed negative correlation.

Table 3. The results of regression (B) estimation

Variables

Within estimation

GLS estimation

Coefficient

P-value

Coefficient

P-value

Constant

0.559

0.472

1.382

0.00

Dummy for macroeconomic shock

0.420

0.023

0.389

0.022

Dummy for specialization level

0.272

0.636

0.054

0.077

Tax income per capita

0.0002

0.649

-0.0001

0.524

R2

0.01

0.11

Breusch-Pagan test for random effects

c2(1)=5.01, P-value=0.03

Hausman specification test

c2(3)=4.78, P-value=0.19

Hausman test shows that we can use GLS estimations. It means that the correlation of the intensity of tax exemptions granting with lobbing pressure and exogenous price levels is statistically significant, however tax level appears to be insignificant.

Table 4. The results of regression (C) estimation

Variables

Within estimation

GLS estimation

Coefficient

P-value

Coefficient

P-value

Constant

11.039

1.731

8.608

0.00

Dummy for macroeconomic shock

4.859

0.001

3.862

0.007

Share of unprofitable firms

-0.017

0.702

-0.080

0.007

Unemployment rate

-0.028

0.210

0.021

0.019

R2

0.02

0.08

Breusch-Pagan test for random effects

c2(1)=1.99, P-value=0.158

Hausman specification test

c2(3)=12.08, P-value=0.007

Data on the interregional trade is not available. Rate of unemployment is used as a proxy for pressure of outside producers on the regional production. Breusch-Pagan test and Hausman test show that we should use within estimations. So the significant is correlation of the price regulation and macroeconomic shocks only. The coefficients of unemployment rate and share of unprofitable firms are statistically insignificant.

So the empirical estimations obtained on the base of aggregated data mostly confirm the comparative static results of the theoretical model. However the structure of the model needs the disaggregated by sectors estimations, these estimations will be presented in the final report.

Plan for further work

Development of the theoretical results

Development of the hypotheses of the empirical analysis

Updating and disaggregation of the empirical information for the empirical analysis

Regression estimations and analysis of the results

Final report, discussion

 

References

In English

Baltagi, B.H. (2001), Econometric Analysis of Panel Data, 2-nd edition, John Wiley & Sons, LTD.

Bernheim, D. and M. Whinston (1986), Menu Auctions, Resource Allocation, and Economic Influence, Quarterly Journal of Economics, vol. 101, N 1, pp. - 31.

Berkowitz, D. (1996), On the persistence of Rationing Following Liberalization: A Theory for Economies in Transition. European Economic Review 40, pp. 1259 - 1279

Bertocchi G. and M. Spagat (1997) Structural uncertainty and subsidy removal for economies in transition, European Economic Review 41, pp. 1709 – 1733.

Boycko, M. (1992), When Higher Incomes Reduce Welfare: Queues, Labor Supply, and Macroeconomic Equilibrium in Socialist Economies. Quarterly Journal of Economics, 107, pp. 907 – 920.

Castanheira, M and V.Popov, (1999), Framework Paper on the Political Economic of Growth in Russia and Central ad Eastern European Countries, mimo.

Dewatripont, M. and G. Roland (1992a), Economic Reform and dynamic political constraints, Review of Economic Studies, 59, pp. 703-730.

Dewatripont, M. and G. Roland (1992b), The virtues of gradualism and legitimacy in the transition to a market economy, Economic Journal 102, pp. 291 – 300.

Drazen A. (2000) Political Economy in Macroeconomics. Princeton University Press.

Edwards, J., M. Keen, (1996) Tax competition and Leviathan, European Economic Review, 40, pp. 113 – 134.

Engel C., J.H. Rogers (1996), How wide is the border? American Economic Review, vol.86, N 5, pp. 1112 - 1124

European Union’s TASIC Program (1996), Analysis of Tendencies of Russian Region’s development in 1992 – 1995. Contract BIS/95/321/057, Moscow, March.

Grossman, G. and E. Helpman (1994) Protection for Sale, American Economic Review, vol. 84, N 4, pp. 833 – 850.

Hilman, A. (1982), Declining Industries and Political-Support Protectionist Motives, American Economic Review, vol. 72, N 5, pp. 1180-1187.

Hilman, A. and H. Ursprung (1988), Domestic Politics, Foreign Interests, and International Trade Policy, American Economic Review, vol. 78, N 4, pp. 729-745.

Hsiao, C. (1986), Analysis of Panel Data. Cambridge University Press.

Kruegel, G., and M. Ciolko, (1998), A Note on Initial Conditions and Liberalization during Transition, Journal of Comparative Economics, 26 (4), pp. 618-634.

Lipton, D. and J. Sachs (1990) Creating a Market Economy in Eastern Europe: The Case of Poland. Brookings Paper on Economic Activity, 1, pp. 75 – 133.

Magee, S., W. Brock and Y. Leslie, Black hole tariffs and endogenous policy theory: Political economy in general equilibrium. Cambridge University Press, 1989.

McKinnon, R. (1991) The Order of Economic Liberalization. Financial Control in the Transition to a Market Economy, The Johns Hopkins University Press.

Murphy, K.M., A.Shleifer, and R.W.Vishny, (1992), “The Transition to a market Economy: Pitfalls of Partial reform,” The Quarterly Journal of Economics, August, pp. 889 – 906.)

Persson, T. and G. Tabellini, (2002) Political Economics: Explaining Economic Policy. The MIT Press.

Polishchuk L. (2000)“Political Economy of Scale and Endogenous Rule of Law”, mimo, IRIS Center, University of Maryland.

Roland G. (2000), Transition and Economics: Politics, Market, and Firms. The MIT Press.

Sah R.K. (1987) Queues, Rations, and Market: Comparisons of Outcomes for the Poor and the Rich, American Economic Review, vol. 77, N 1, pp. 69 –77

Shliefer, A. and R. Vishny, (1992), Pervasive shortages under socialism, Rand Journal of Economics 23 N 2, pp. 237 –246.

Stigler ,G. (1971), The Theory of Economic Regulation, Bell Journal of Economics, vol. 2, N 1, pp. 359 – 365.

In Russian

Áåðêîâèö Ä., Äåéîíã Ä. (2000), Ãðàíèöà âíóòðè Ðîññèéñêîãî ýêîíîìè÷åñêîãî ïðîñòðàíñòâà. Ðåãèîí: ýêîíîìèêà è ñîöèîëîãèÿ, ¹ 1, c. 85-99.

Âåðõîâåíñòâî Êîíñòèòóöèè ÐÔ è ôåäåðàëüíûõ çàêîíîâ – îñíîâíîé ïðàâîâîé ïðèíöèï (2000), Õîçÿéñòâî è ïðàâî, ¹ 1, ñ. 32 – 41

Ãëóùåíêî Ê.Ï, (2001), Ïðîñòðàíñòâåííîå ïîâåäåíèå óðîâíåé öåí, ÝÌÌ, ò.37, ¹  3, ñ. 3 –13

Äîêëàä Âñåìèðíîãî áàíêà «Ðàçðóøåíèå ñèñòåìû íåïëàòåæåé â Ðîññèè: ñîçäàíèå óñòîé÷èâîãî ýêîíîìè÷åñêîãî ðîñòà», (2000), Âîïðîñû ýêîíîìèêè, ¹ 3, ñ. 4 – 45.

Êîëîìàê Å. (2001), Áþäæåòíàÿ äåöåíòðàëèçàöèÿ è ìÿãêèå áþäæåòíûå îãðàíè÷åíèÿ ïðåäïðèÿòèé. Ýêîíîìè÷åñêîå ðàçâèòèå Ðîññèè: ðåãèîíàëüíûé è îòðàñëåâîé àñïåêòû, âûïóñê 2, ÑÎ ÐÀÍ, ÈÝÎÏÏ.

Õåíñîí Ô. (2001) Âëèÿíèå ôàêòîðà ðåãèîíàëüíîãî ðàçíîîáðàçèÿ íà ýêîíîìè÷åñêóþ òðàíñôîðìàöèþ Ðîññèè, Ïðîáëåìû ïðîãíîçèðîâàíèÿ, ¹ 3

 


Appendix 1

Goods and services subject to price regulation at regional level in Russia

Housing

Land

Monopolized sectors: oil processing, metallurgy, engineering, chemical, petrochemical, wood-processing industry, pulp and paper, textile, food, extracting, timber, non-ferrous metallurgy, construction materials

Agriculture products

Public transport

Construction

Electric power

Heating

Medical services

Service of municipal firms and organizations

Educational services

Architecture and city designing

Social services

Culture service

Gas natural and liquefied, solid fuel

Glass

Grain, alcohol, sugar, salt, oil, flour, potatoes, children's meal, meat, sausage, eggs, matches, bread, milk and dairy production, pasta, butter, fish, canned food, dry milk, tea, vegetables, synthetic detergents, soap.

Water-supply and sewage

 

Source: Legislative data base «Consultant Plus. Regional Legislation»,


Appendix 2

Number of documents on price control adopted in regions of Russia

 

 

1992

1993

1994

1995

1996

1997

1998

Republic Adigeia

 

 

1

0

1

1

4

Republic Altai

 

 

 

 

 

 

14

Altaiskei krai

 

 

4

19

48

9

32

Amurskya oblast

 

 

25

2

1

6

16

Arkhangelskaya oblast

 

 

 

 

3

2

22

Astrakhanskya oblast

 

 

 

18

3

2

2

Republic Bashkortostan

 

 

2

35

4

1

3

Belgorodskaya oblast

4

15

23

35

9

6

24

Bryanskaya oblast

 

 

 

 

1

 

1

Republic Buryatiya

 

 

 

 

17

15

7

Vladimirskaya oblast

11

1

1

18

 

1

21

Volgogradskaya oblast

 

 

 

2

1

 

 

Vologodskaya oblast

1

5

2

2

5

8

11

Voronezhskaya oblast

 

34

15

20

8

8

17

Republic Dagestan

 

 

 

 

 

 

3

Ivanovskaya oblast

6

2

2

2

19

2

17

Irkuskaya oblast

 

 

 

16

2

1

6

KabardinoBalkarskaya Republic

 

 

1

 

 

 

15

Kaliningradskaya oblast

 

1

2

4

1

1

26

Republic Kalmikiya

 

 

 

 

 

 

12

Kaluzhskaya oblast

 

 

2

22

3

1

1

Relublic Kareliya

 

 

 

 

 

3

1

Kemerovskaya oblast

 

2

 

3

4

2

17

Kirovskaya oblast

16

 

 

16

1

11

19

Republic Komi

10

1

2

4

9

10

23

Kostromskaya oblast

 

 

 

1

2

2

1

Krasnodarskei krai

 

5

7

22

5

6

28

Krasnoyarskei krai

 

 

2

18

4

15

21

Kurganskaia oblast

 

 

 

 

17

4

31

Kurskaya oblast

 

 

1

1

4

15

19

Lipetskaya oblast

 

 

 

5

6

14

27

Republic Marei –El

 

 

 

40

6

1

1

Moscow

7

24

7

37

 

5

23

Moskovskaya oblast

25

 

1

25

 

 

 

Republic Mordiviya

1

6

8

2

35

3

29

Murmanskaya oblast

3

7

13

5

3

19

1

Nenetskei ÀÎ

 

 

 

 

 

1

17

Nizhegorodskaya oblast

26

26

24

27

3

4

2

Novgorodskaya oblast

 

 

4

 

 

 

 

Novosibirskaya oblast

 

2

9

22

8

16

26

Omskaya oblast

 

 

3

2

18

2

19

Orenburgskaya oblast

 

 

 

 

11

 

18

Orlovskaya oblast

11

 

1

 

 

 

 

Penzenskaya oblast

 

1

21

22

6

1

20

Permskaya oblast

 

 

 

16

 

2

8

Primorskei krai

 

1

4

8

11

21

30

Pskovskaya oblast

 

 

 

2

7

 

6

Rostovskaya oblast

24

22

1

16

4

16

26

Pyazanskaya oblast

 

15

11

2

4

 

26

Samarskaya oblast

8

 

1

1

3

5

 

Saratovskaya oblast

 

 

 

 

 

1

11

Republik Sakha (Yakutiya)

 

 

 

 

 

 

29

Sakhalinskaya oblast

 

17

1

17

12

5

31

SanktPetersburg and Leningradskaya oblast

33

1

3

5

2

3

13

Sverdlovskaya oblast

33

24

12

7

35

3

19

Smolenskaya oblast

 

6

8

1

2

3

25

Stavropolskei krai

 

1

1

1

6

4

4

Tambovskaya oblast

 

 

0

2

4

29

17

Republic Tatarstan

1

2

12

26

7

5

10

Tverskaya oblast

 

 

 

4

 

2

22

Tomskaya oblast

 

2

7

12

2

4

14

Tulskaya oblast

 

 

1

26

5

2

3

Tumenskaya oblast

 

 

 

3

2

1

15

Republic Udmurtiya

7

11

1

20

0

2

35

Ulianovskaya oblast

1

 

6

29

1

 

39

Khabarovskei krai

 

23

17

6

40

13

20

Republic Khakasiya

 

 

 

15

 

2

8

Khanti-Manseiskei ÀÎ

 

 

 

3

17

23

38

Chelyabinskaya oblast

21

18

12

15

16

7

6

Chitinskaya oblast

 

 

 

 

1

1

 

Republic Chuvashskaya

4

2

1

 

1

4

11

Yaroslavskaya oblast

6

1

5

8

7

8

7

Source: Legislative data base «Consultant Plus. Regional Legislation»,

 

 


Appendix 3

Descriptive statistics and preliminary estimations

Table 1

Subsidies to enterprises from federal and regional budgets, percent of GNP

 

1992

1993

1994

1995

1996

1997

1998

Subsidies from federal budget

5,8

2,5

3,1

2,2

1,6

1,8

1,9

Subsidies from the regional budgets

5,3

6,8

7,3

5,2

6,3

6,9

7,2

Source: Russian Statistical Yearbook, 1999

Table 2

Regional subsidies and documents regulating price

 

1992

1993

1994

1995

1996

1997

1998

1999

Share of the regional subsidies in the budget expenditures (average level for Russia)

 

-

 

-

 

-

 

-

 

18,5

 

18,3

 

21,9

 

18,8

Number of documents adopted at regional level in Russia directed on price regulation, total

 

312

 

321

 

288

 

692

 

457

 

364

 

1070

 

-

Source: Legislative data base «Consultant Plus. Regional Legislation», data of Ministry of Finance RF

 

Table 3.

Summary statistics on the regional subsidies,(share of budget expenditures, percentage

 

Total subsidies

Subsidies to industry, agriculture and construction

 

1996

1997

1998

1999

1996

1997

1998

1999

Minimum

0,0

0,0

0,0

0,0

0,0

0,0

0,0

0,0

Maximum

50,0

48,5

44,3

47,5

11,6

14,2

13,7

8,7

Median

18,9

19,0

22,8

19,1

3,9

3,6

3,9

2,8

Average

18,5

18,3

21,9

18,8

4,2

4,1

4,0

3,0

Standard deviation

10,8

9,7

9,3

7,9

3,1

3,2

2,6

2,0

Source: data of Ministry of Finance RF

 

 

 

 

 

 

Table 4

Branch structure of the regional subsidies, percentage

 

1996

1997

1998

1999

Industry, construction and power engineering

8,4

2,9

4,8

3,4

Agriculture and fishing

10,9

12,8

9,9

10,0

Protection of environment

0,7

0,3

0,6

0,9

Transport, road construction, communication

15,9

12,6

15,3

14,1

Market infrastructure

0,1

0,3

0,3

0,2

Housing

63,2

68,9

69,1

66,4

Others

0,7

2,2

0,0

4,9

Source: data of Ministry of Finance RF

 

Table 5

Regions granting tax exemptions, percentage

 

1992

1993

1994

1995

1996

1997

1998

Tax relief for particular enterprises

1

3

8

17

26

26

28

Tax relief for branches of industry

1

6

14

25

38

40

47

Tax relief for small business

0

1

3

4

11

14

19

Free economic zones

1

1

3

6

7

10

11

Tax relief for investors

1

3

4

11

22

51

79

Source: Legislative data base «Consultant Plus. Regional Legislation»

Table 6

Estimations of the regression of subsidies on price regulation

 

Fixed effect model

Random effect model

 

a

P-value

a

P-value

Number of documents adopted at regional level regulating prices and tariffs

 

0.12

 

0.028

 

0.11

 

0.022

 

 

Table 7

Estimations of correlation between subsidies and transfers

 

Fixed effect model

Random effect model

Coefficient

P-value

Coefficient

P-value

VAT transfers

0.266

0.000

0.217

0.000

«Pure» transfers

-0.200

0.136

-0.223

0.090

 

 

Table 8

Estimations of correlation between tax exemptions and transfers

 

Fixed effect model

Random effect model

Coefficient

P-value

Coefficient

P-value

Tax relief for particular enterprises

VAT transfers

0.007

0.451

0.013

0.065

«Pure» transfers

-0.004

0.450

0.007

0.057

Tax relief for branches of industry

VAT transfers

-0.018

0.047

-0.024

0.001

«Pure» transfers

0.013

0.011

0.008

0.040

 

 

Table 9

Variation of price indicators in Russia, percentage

 

1992

1993

1994

1995

1996

1997

Index of consumer prices

26,1

22,3

8,9

6,7

3,6

10,1

Index of electric-power prices

43,5

33,5

24,8

19,9

19,0

17,8

Value of consumer good basket, consisting of 25 food products

-

33,7

30,2

33,6

35,7

47,3

Share of value of the consumer good basket into money income of population

-

31,8

27,7

44,2

44,1

49,6

Source: Prices in Russia. Statistical Yearbook: Goskomstat of Russia, M.-1998.