Since the restoration of an independent state budget Estonia has pursued the policy of balancing revenue and expenditures. The foreign loans taken by the Government as well as domestic loans taken by local governments have been used for investments or renovations; rather than for consumption. Disregarding state loans as well as loans guaranteed by the state and on-lent to production organisations, revenue and expenditures in the fiscal system(*) are balanced in 1994-1995. In addition to that, the Estonian society seems to have come to a tacit agreement that taxes must not be too high.
(*)Unless specified otherwise, by the fiscal system we mean the sum total of the state and local budgets plus money in the social insurance and health insurance funds.
The tax level is usually measured as a ratio of tax revenue to the gross domestic product (GDP). According to analysts, the wealthier the society, the higher the so-called normal tax burden is. Wealth is measured by per capita GDP. Under this approach, Estonia's 35-40% tax burden is a bit too high for the country's level of development, but too low for affluent industrial countries such as Sweden or the Netherlands. Compared to other transition economies, Estonia can still be placed among the countries with a modest tax burden(**) (that is, below 40% of the GDP). Estonia is exceptional in the sense that the relative reduction in the large amount of money being pumped through the budget system has not been followed by a spate of tax increases (at the end of the 1980s an estimated 43-44% of the GDP was redistributed through the budget system).
(**)Due to methodological difficulties in measuring the GDP indices of the tax load differ in publications of different institutions, but the listings of countries by tax burden are usually similar.
In a phenomenon characteristic of the Central European countries, a year or two after the beginning of the transition period the share of taxes in GDP begins to fall and the problem of balancing revenue and expenditures becomes more acute. In Estonia, initially the corporate income tax based on inflational profits resulting from abolishing price controls simply masked the inefficiency of tax collection and the resultant fall of revenue. In order to compensate for the reduction of taxes paid from profits, the governments of several countries have been forced to increase taxes.
Since monetary reform, the political will to change the tax rates in Estonia has most clearly been expressed in 1993 during the adoption of new tax laws (laws on tax collection, state budget, income tax, local budgets, VAT, the law on relations between the central and local budgets). The new Income Tax Law considerably reduced the size of the tax (the corporate income tax was reduced to 26% and the 33% personal income tax was replaced by a flat proportional 26% tax). The current structure of tax revenues is close to the average of the countries of the Organisation for Economic Cooperation and Development (OECD) (see Figure 1). The only exception is the small share of customs duties due to Estonia's liberal trade policy (see Table 1).
In Estonia, the period beginning in 1994 can be regarded as a stage of fine adjustments of the taxation system. Despite the new tax laws adopted in 1993, which have been in force since the beginning of 1994, the tax burden has changed surprisingly little in recent years (in 1993 it was 37.5%, in 1994, 36.9%, and for 1995, an estimated 37.5-38.0%). Besides the changes in legislation, the tax burden was affected just as much by changes in the structure of the tax basis. The tax burden of 1994, however, differed very little from that of 1993 because:
In general, the above factors should have cancelled each other out and therefore, we can say that the fall of the tax burden in 1994 stemmed from the increase in tax debts (debts from unpaid taxes alone increased by 0.7% against the GDP).
A new tendency since 1994-1995 has been the increasing excise tax burden due to adjustments (unavoidable because of inflation) and the introduction of new excise taxes: in 1993 the share of excise taxes in the GDP was 1.8%, increasing to 2% in 1994 and estimated to reach 2.3% this year. It has also been said that the reduction of state revenue due to increases in the tax-free minimum would be compensated by increasing excise taxes on alcohol and fuel. According to the 1996 draft budget, revenue from excise taxes should double as compared to the revenue expected in 1995, thus raising the share of excise taxes in GDP to 4%. Compared to the majority of other countries this is high (it must also be taken into account that in some countries differentiated VAT plays the role of the excise tax).(***)
(***)In East European countries and in the former Soviet Union excise taxes in the period 1986-1992 on an average amounted to 2.2% of the GDP, the respective figure for countries of the OECD was 2.9%.
The Estonian tax system with its neutral character and overwhelmingly proportional tax rates should automatically guarantee a stable tax burden in the years ahead, provided that taxe rates and the structure of GDP are not changed. The stability of the tax system is, of course, an important prerequisite for attracting foreign capital.
The adoption of the above-mentioned tax laws in combination with the developments in the structure of the tax base have resulted in considerable changes in the structure of tax revenue (see Figure 2). While in 1994 the tax revenue was influenced mostly by changes in the tax system, then this year's changes in the structure of revenue already stem from changes in the structure of the tax base. An exception here is the above-mentioned policy on excise taxes and the freezing of the tax-free minimum of private individuals on the level of the beginning of 1994, which means increasing the actual tax level.
The same can be said about the distribution of tax revenue between the different parts of the fiscal system. The 1993 decision to reduce the share of personal income tax going into local budgets to 52% instead of the former 100%, reduced the share of local governments in the total state revenue from 24% in 1993 to 12.5% in 1994 (see Table 2). The year's-end also marked the reassignment of functions between the state and local budgets. The main obligation transferred to the state budget was that of paying salaries to teachers of general-education schools. This also reduced the share of expenditures by local governments.
At the beginning of 1995 no such redistribution of revenue took place.
In 1996, the share of revenue from personal income tax paid to local governments is planned increase from 52% to 66%. The task of paying housing and subsistence subsidies is planned to deliver to local governments. Such a change in relations with local budgets can be called administrative decentralisation (a reverse process took place in 1994).
Delegation of responsibilities together with a corresponding amount of revenue is fairly common in transition economies. The aim is to reduce expenditures in the budget system as a whole. It is characteristic of Estonia, too, that at negotiations between central and local governments the local bodies have usually estimated the costs of fulfilling responsibilities laid on them at 10-12% higher than the financing proposed by the central government. Local governments have been conservative in taking advantage of the possibilities offered in the 1994 Law on Local Taxes since the costs of tax collection are relatively high. It is estimated that in 1995, the average revenue from local taxes will not be higher than 0.3-0.4% of the total revenue of local governments. Only in Tallinn might the share of local taxes exceed 4%.
The increase of the share of local governments' revenue this year implies the continuing increase of the share of the salaries and wages fund in the GDP. Last year, the share of wages and salaries in the GDP rose to 41.1% (38.7% in 1993). In other words, the distribution of revenue within the fiscal system changes considerably due to changes in the structure of the tax basis, that is, regardless of the Government policy. In the period from 1994-1995 these changes can be epitomised by the ratio of VAT to the personal income tax. It is the rate of collection of these two taxes that determines whether the state budget is under more duress in regard to planned expenditures than other parts of the fiscal system. In 1996 the tensions would hopefully already be eased by the Treasury.
In principle, revenue from the VAT and (personal) income tax, if the rates are kept stable, should change at a fairly similar rate, since the salaries and wages fund is the biggest unit of the value added tax base. In the years 1993-1994 these two taxes were collected at a different rate. However, differences in the rate of increase seem to have been diminishing over the last nine months. As compared to the previous year, 1994 can be called "the year of the VAT" and 1995 seems to become "the year of the personal income tax" (see Figure 3and Figure 4).
The different increase of the revenue from these two taxes has two reasons. Firstly, a number of important services (rent, heating for inhabitants) are exempt from VAT, while at the same time changes in the price of these services are the major factor affecting the size of wages and salaries. A second reason is the high share of foreign trade in the GDP (117% in 1994, for example). Both these aspects indicate that the make-up of tax revenue is much affected by foreign trade. By the middle of this year the increase of foreign trade turnover seemed to have come to a halt and its ratio to the GDP could have been estimated at 116-118%. This allows us to forecast small changes in the relations of the tax revenue growth, unless foreign trade receives some additional outside impetus for development (such as increased demand by neighbouring countries, increased foreign financing, technical development of the infrastructure of the transit trade, etc). That in the 1996 draft state budget the emphasis is on the increase of VAT, suggests that a higher increase rate of the foreign trade is expected. Thus, in 1995-1996 changes in the structure of tax revenue are first and foremost the passive reflection of the relations of elements of the GDP.
The great impact of foreign trade on state revenue poses no economic problem, but can cause some technical difficulties for predicting the development of the fiscal system. Thus, in 1996, 51% of state revenue is expected from value added tax. However, if in reality wages increase more than expected and foreign trade does not increase more rapidly than domestic trade, it would mean that the share of the state budget would decrease in the total revenue of the fiscal system. Since the share of income tax and VAT in the revenue of the fiscal system is high (nearly 75%), this will help to perpetuate the existing share of revenue in the GDP, provided tax rates remain unchanged. However, this can have a destabilising effect on the relations of the parts of the fiscal system.
The distribution of revenue within the fiscal system can also be more influenced by differences in price increases in the sheltered and open sectors of the economy. The rise of the average wage and revenue from personal income tax depend more on price increases in the former, while revenue from VAT depends more on price increases in the latter. Since VAT is the biggest source of income for the state budget, state revenue is more affected by price increases in the open sector than by increases in the sheltered sector.
The first months of this year have been exceptional in the sense that all current revenue has been used for current expenditures. In earlier years (1993-1994) it has been possible, in the first months of the year, to save money for an additional budget and there has been no problem balancing revenue and expenditures during periods shorter than the fiscal year. Last year, for example, the surplus of state revenue in the first quarter amounted to 16.1%; in the second and third quarters, the surplus decreased and the savings were spent in the fourth quarter which, in the sense of the balance of current revenue and expenditures, ended with a 15.4% deficit. In 1995, revenue has outstripped expenditures by 4.5%.
Above we referred to the stability of tax revenue in 1993-1995. In the view of this, the "disappearance" of surplus revenue and additional budgets is a purely technical matter characterising more the planning of revenue and expenditures than the potential revenue. In essence, there is no difference whether to change the originally planned structure of expenditures (which is obviously inevitable also in the future) through drawing up additional budgets or through changing the law on the state budget. The year 1995 is a so-called transition year in a sense that the proportions of the state budget are adjusted through amending the budget law, and using both additional revenue and the surplus revenue of the previous year.(****)
(****)On May 17 Riigikogu amended the Law on the 1995 State Budget and redistributed money between ministries, without increasing the size of the budget. In July the excise tax on alcohol was increased and in anticipation of higher revenue a Law on the First Additional Budget was adopted, increasing subsidies to local governments. The adoption of the second additional budget was planned for the autumn, using money from the surplus revenue from the previous year. This way, expenditures can be specified within the range of 1.5-2.0%, which is incomparable to the size of additional budgets in 1994 (17%). Due to the above-mentioned different increase rates in the VAT and the income tax, more revenue than expected has flown the local governments and the health insurance fund this year. Therefore, these two can also pursue the policy of the so-called traditional additional budgets.
A balanced budget policy together with a stability-oriented tax system are regarded highly restricting factors from the point of the state expenditures. In this context, the increasing share of the public sector in the consumption side of the GDP may seem controversial. Yet, the increase has been substantial, from 16% in 1993 to 21.4% in 1994.
As the turnover of foreign trade in 1993-1994 increased more than the GDP, the changes are more noticeable in domestic demand. As we can see, the share of both investments and the end consumption of the public sector have increased while private consumption has decreased (see Table 3).
Although the tax revenue remained stable in 1993-1994, revenue from ownership and other (non-tax) sources increased from 0.8% of the GDP in 1993 to 2.7% in 1994. This allowed to increase expenditures from 38.3% of the GDP to 39.9%. Such an increase of expenditures does not necessarily have to lead to the increase of the end consumption of the public sector by the logic of accounts of the national economy. The structure of the expenditures, too, has to change towards the end consumption.
While the size of expenditures and the principles of balancing revenue and expenditures are determined at the most general level of budget policy then at the next level decisions have to be made the structure of expenditures by their economic content. Due to the laws governing the economy, the goverment sector must also maintain certain proportions between the different kinds of expenditures (wages, investments, compensations, subsidies, etc.).
There has been much discussion in Estonia about the balanced budget policy and different political forces have approved the principles of this policy. However, the structure of expenditures by economic purpose has drawn little attention and has developed mostly by itself. The only exception, ever since the restoration of independence, has been the general negative attitude towards subsidies and their replacement by compensations to people who need financial support. After replacing heating subsidies in 1994 with housing benefits we can speak only of subsidisation of public transport. The share of subsidies has declined below 1% of the GDP.
In a transition economy, the highest priority expenditure is wages and the lowest priority is attached to investments. Cutting the latter is the most widely used means of reducing expenditure and "cheap" in the political sense since it poses no direct threat to the existence of any official post or government department. Usually, the share of investments in the budget system amounts to approximately 2% of the GDP in transition economies. In the developed industrial countries the figure is 2.0-2.5 times bigger.
In Estonia, the share of public sector investments after 1991 has been between 1% and 2% of the GDP. In 1994, for example, such investments (without the cost of repairs) amounted to only 1.3% (see Table 4). In order to halt the reduction of investments, a programme of state investment for the next three years was drafted last year. From the financial aspect this programme is divided into two parts, one financed from revenue from taxes and the other from foreign loans. The latter includes money on-lent by the Government. Drafting of the first programme, for the years 1995-1997, resulted in an increase in the share of investments. The next programme (for 1996-1998) is unlikely to cause a similar increase since the size of investments financed from tax revenue is scheduled to increase only 1.27 times from the level of 1995, and the share of public sector investments is not much over 2% of the GDP. The combined investments from both sources into social and technical infrastructure under the programme of state investments are to be maintained around 4.5% of the GDP.
The programme has come under criticism from some economists since the bulk of the money is to be spent on repairs and small projects which does not allow to concentrate on long-term economic and political goals.
The share of expenditures on labour (without the social tax) has been constantly increasing in the state budget (16.7% in 1993; 18.1% in 1994; an estimated 21.6% in 1995) although this has not been the case in the entire fiscal system. The increase of expenditures can result from both the increase in the number of workers as well as the increase of the average wage. This leads to the supposition that in government ministries and departments the number of employees and their average salary has increased more rapidly than in other branches of the fiscal system. The money allocated in the state budget on salaries accounted for 4.1% to 4.9% of the GDP in 1994-1995. According to a study of the International Monetary Fund, based on data of the 1980s, the average share of expenditures on wages and salaries of the central government in industrially developed countries was 5.6% of the GDP, and 7.7% of the GDP in developing countries. However, since the division of the units of the fiscal system differs from country to country we cannot draw very far-reaching conclusions from this.
If we compare the ratio of the salaries and wages fund of the entire fiscal system and the GDP, the result is analogous. Considering the Government's promise to increase the salaries of teachers by one-third in the last months of this year, the average wage in 1995 should amount to 9% of the GDP. According to the above-mentioned IMF study, this figure was 11% in industrially developed countries, and 6.8% in the developing countries.
It does not follow from the above as if Estonia had unlimited possibilities to increase wages and salaries through cutting other expenditures. Besides investments, compensations to the population are another category of expenditures that cannot be reduced. Over the last few years, compensations have amounted to approximately 9% of the GDP (see Table 4 and Table 5). Pensions amount to 6.5%-6.6% of the GDP. The source of the pensions, according to a law adopted in 1994, is the social tax paid from the salaries and wages fund, and which cannot be used for any other purpose.
If we proceed from the thesis that the current tax level is normal, or even too high, the share of wages of the fiscal system in the GDP may also begin to increase, although the GDP itself is lower than that in the developed countries. This is confirmed by the comparison of the share of labour costs in different countries. In Estonia, wages and salaries make up 19-20% of the total expenditures in the state budget. The IMF study put the respective figure of industrially developed countries at 15% and that of the developing countries at 27%. The increase of the labour costs has resulted from the increase of the so-called other costs and in the near future we have two options:
The distribution of budget expenditures by spheres of activity is the most-debated area in all discussions about the state budget, yet no party-programme priorities have been followed.
Since 1991, the basic means of balancing revenue and expenditures in drafting the budget has been proportional cuts of expenditures in all spheres. The period from 1991 to 1992 can be called the period of restoring Estonia's independent economy. After that, expenditures by category have been relatively stable (see Table 6).
As usual for small countries, priority is given to expenditures on the social sphere. It would be premature to look for any economic and political intentions in the relatively small fluctuations of these expenditures in different years. Rather, one could suppose that the more limited the resources were, the higher the share allocated to the social sphere would be. The reduction of expenditures on the social sphere to 48.4% in 1994, for example, refers rather to the fact in the previous year when the share of additional revenue (additional budgets) as compared to the main budget was higher and thus enabled additional financing of other spheres as well.
While we can consider the share of expenditures on the social sphere as constant, then in case of other spheres it becomes clearly evident that expenditures on areas involved with the restoration of state independence (administrative, law enforcement, justice, defence) have increased compared to the expenditures on economic development. The same tendency has continued in the first half of this year. The Government has declared law enforcement, justice, and export subsidies to agriculture the priority areas in the 1996 state budget. In financial terms, however, these priorities are of a lesser value than, for example, the wage policy of the social sphere. Therefore, the 1996 structure of expenditures by spheres of activity differs little from that of the previous years.Andres Saarniit