Slovenia
Recent Economic Developments
Author:
International Monetary Fund
Search for other papers by International Monetary Fund in
Current site
Google Scholar
Close

This paper reviews economic developments in Slovenia during 1990–96. Slovenia experienced its first positive real GDP growth in 1993. Real GDP grew by 1.3 percent. This modest recovery began under the impetus of buoyant domestic demand, which grew by 8¼ percent; real foreign demand contracted by 6½ percent owing to a recession in Western markets. Despite the growth in real aggregate demand by more than 2 percent, the output response was dampened as domestic demand growth spilled primarily into a boom in consumer goods imports.

Abstract

This paper reviews economic developments in Slovenia during 1990–96. Slovenia experienced its first positive real GDP growth in 1993. Real GDP grew by 1.3 percent. This modest recovery began under the impetus of buoyant domestic demand, which grew by 8¼ percent; real foreign demand contracted by 6½ percent owing to a recession in Western markets. Despite the growth in real aggregate demand by more than 2 percent, the output response was dampened as domestic demand growth spilled primarily into a boom in consumer goods imports.

I. Real Sector

1. Overview

Contractionary forces stemming from the disintegration of the former SFRY and the CMEA as well as the move to a more market-oriented system had induced a cumulative fall of 12 ½ percent in real GDP in Slovenia over the period 1990-91. Output fell by another 5 ½ percent in 1992—the first full year of independence—due to weak domestic demand and lingering effects of the loss of markets in the former SFRY.

Slovenia experienced its first positive real GDP growth in 1993 (Table 1 and Chart 1). Real GDP grew by 1.3 percent. This modest recovery began under the impetus of buoyant domestic demand which grew by 8 ¼ percent; real foreign demand contracted by 6 ½ percent due to a recession in Western markets. Despite the growth in real aggregate demand by over 2 percent, the output response was dampened as domestic demand growth spilled primarily into a boom in consumer goods imports. On the supply side, the recovery in real GDP was entirely on account of the service sector which grew by 4 percent.

Table 1.

Aggregate Demand and Supply

article image
Sources: Slovenian authorities; and staff calculations.
CHART 1
CHART 1

SLOVENIA: REAL ECONOMIC ACTIVITY

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by Slovenian authorities.1/ The series are on income and expenditure for socially- and state-owned enterprises, and beginning January 1992 includes private enterprises (with more than 3 employees).

Real GDP growth accelerated to 5 ½ percent in 1994, was broad-based and sectorally more balanced than in 1993. The primary stimulus came from a rapid expansion in investment as uncertainties associated with the program for privatization of socially-owned enterprises (SOEs) receded, and merchandise export growth became positive as economic activity recovered in Europe. Earnings from tourism also expanded. Consumption demand was moderated by slower real wage growth and higher savings rates; aggregate demand expansion channeled itself strongly into real GDP growth.

2. Aggregate demand

Over the period 1992–94, aggregate demand witnessed a qualitative and quantitative change, first turning positive in 1993. The contribution of domestic and foreign components shifted in response to domestic and external conditions. Real wage growth had begun to exert expansionary pressures on private consumption demand as early as the first quarter of 1992. These pressures were especially manifest in 1993, but receded somewhat in 1994. A recovery in Western markets began in mid-1993, and registered its full impact on aggregate demand only in 1994. In the meantime, along with domestic demand expansion, export growth had created the conditions for a revival of investment expenditures.

Real aggregate demand contracted sharply in 1992 falling by 19 percent. Merchandise exports were the main source of the contraction. Private consumption expenditures also fell by 3 ¼ percent as real disposable income declined by 4 percent. The decline in income was caused by: (i) a decline in real average wage by 9 percent; (ii) a decline in total employment by 6 ½ percent; and (iii) a spillover effect on wage-indexed social transfers. Investment activity recovered only somewhat, due to the uncertainties surrounding the future of the privatization program, rising by only 5 percent. A survey of industrial enterprises reported that only 19 percent of firms indicated that credit availability was satisfactory, suggesting that financial constraints may also have been a factor dampening investment activity, in particular for newly created firms. Net capital formation remained negative for a third consecutive year.

Aggregate demand growth turned positive in 1993. The real average wage grew by 16 ½ percent in 1993 over 1992. The impact on disposable income, which rose by 7 percent, was moderated by a continuing decline in total employment. The greater purchasing power of the household sector translated into a 5 ¾ percent increase in private consumption expenditures. Gross investment growth also doubled to 11 ½ percent and net investment became slightly positive. A substantial portion of the increased consumption expenditures were, however, satiated by a rapid expansion in imports of goods (and services) which grew by 3 ½ percent. In addition, earnings from merchandise exports was negative and overwhelmed a rebound in tourism earnings; exports of GNFS fell by 6 ½ percent (see Chapter II on External Sector Developments).

Aggregate demand growth rose to 4 ¼ percent in 1994, helped by robust growth in all components. Investment expenditures rose by 15 percent as the privatization process gathered momentum and there was greater certainty about the ownership transformation program (see Chapter V on Structural Reforms). The recovery in Western markets provided a boost to merchandise exports which, together with the exports of non-factor services (primarily tourism and cross-country transportation services), recorded a growth rate of 3 ¾ percent. Private consumption demand growth slowed down to 3 ¼ percent as real wage growth was moderated by a restrictive income policy introduced in mid-1994.

The structure of household income also displayed a changing character during 1992–94. 1/ The share of wages and salaries, non-wage income and social transfers in total household income had averaged 55 percent, 17 percent and 28 percent, respectively in 1990–91. Over 1992–94, the share of wages and salaries fell to 50 percent, as the share of non-wage incomes and social transfers rose to 20 percent and 30 percent, respectively. Despite rapid growth in 1993, the level of wages and salaries was still 20 percent below its level in 1990, while other receipts were 26 percent higher and social transfers 18 percent higher than their respective levels in 1990. This partly reflected the downsizing of the work force experienced during the period and the aging of population, plus early retirement and tax breaks on the non-wage incomes (see the section on Labor Markets).

3. Sectoral composition of GDP and recent trends

Slovenia has a diversified productive structure with a high share of industry and services in value added averaging 39 and 55 percent, respectively over 1992–94 (Table 17). As a part of the former SFRY, Slovenia was the source for a large part of the manufactured output of the domestic economy as well as the exit point for exports to Western markets. In part, due to the disintegration of the internal market, there has been a gradual shift from heavy dominance of industry toward a higher share of services in GDP. 1/

a. Industry

Industrial production in Slovenia began on a downward trend during the late eighties as the economy of former SFRY faltered (Table 24). The decline continued until the middle of 1993 when a recovery, initiated by rising domestic demand, began. However, industrial production, which had fallen by over 13 percent in 1992, continued to contract in 1993, declining by 3 percent. This decline was broad-based with all three major branches of industrial activity experiencing a contraction: manufacturing output fell by 2 percent, electricity, gas and water supply by 3 percent, and construction activity by 4 percent. The turnaround in 1994 was decisive and strong. Real growth in these three branches was 7 percent, 6 ½ percent and 7 ½ percent, respectively, propelling total industrial output growth to 6 ½ percent.

The successful reorientation to the European markets, strong demand in partner countries and recovery in domestic investment contributed to the industrial expansion. The commodity composition of industrial growth reflected these factors. While the growth in final goods had already turned positive in 1993, the growth rate was, however, less than 1 percent, as a large share of private consumption demand was satisfied by imports. The recovery in capital and intermediate goods production came only in 1994 with growth rates of 13 ½ and 8 ¾ percent, respectively.

b. Services

The service sector was the first to register a positive growth rate, expanding by 4 percent in 1993. The growth moderated to 3 ½ in 1994. The share of services in value added increased from 54 percent in 1992 to 56 percent in 1994. Rapid growth in the public sector in 1992 and 1993 was a contributing factor to this expansion of the service sector. The private sector is also developing rapidly and represented 88 percent of the total number of firms, 16 percent of total employment (up from 2 percent in 1990) and 24 percent of total revenues in 1994 (Tables 3032).

Tourism, as measured by the number of nights spent by tourists, fell by almost half between 1989 and 1991, reflecting the uncertain political and security situation (Table 26). As the security environment in the region improved, tourism recovered; overnight stays by tourists in 1993 and 1994 increased by 5 ½ and 9 percent, respectively. Despite this recovery, the number of overnight stays was still 26 percent below the peak level recorded in 1990. Tourism-related industries also showed good results. Hotels and restaurants grew by 6 percent in 1993 and 8 percent in 1994, transport by 3 percent in 1993 and 6 percent in 1994.

Within the domestic market, retail turnover grew by 3 ¼ percent in 1993 and 7 percent in 1994. The share of wholesale and retail trade, repairs services remained at about 10 percent of GDP. Small, private enterprises thrived in retail trade, and had better business results than large socially-owned firms. Financial and business-related services increased by 3 percent in 1993 and 2 ½ percent in 1994.

c. Agriculture, forestry and fishing

Agriculture accounts for about 5 percent of GDP. After continued declines of 6 percent in 1992 and 3 ¾ percent in 1993, it grew by 6 ½ percent in 1994. The private sector is playing an increasingly important role in this sector, and produced 75 percent of the value added in agricultural production in 1994 as compared to 66 percent in 1989.

4. Labor market developments

In 1988, socially-owned enterprises accounted for 85 percent of total employment, and agriculture accounted for approximately half of private employment (Vodopivec and Hribar-Milic (1993)). In socially-owned firms job security was guaranteed and labor shedding was banned by law. As a corollary, hidden unemployment—workers officially employed but not performing any meaningful work—reached 13 ½ percent in 1988 (Mencinger (1989)). Other important characteristics of the pre-independence labor market include: (i) a sectoral allocation of resources tilted towards heavy industries to the disadvantage of service industries, and (ii) a relatively high labor force participation rate, in particular for women.

After 1990, total employment shrank rapidly while unemployment reached 15 percent of the labor force (Chart 2 and Table 33). This was accompanied by a reallocation of labor from large enterprises to the public sector and small firms or the self-employed. Real wages also recorded a precipitous decline during the initial years of the transition (Table 34). This adjustment, however, was short-lived and by 1994, the real average net wage was back to its 1990 level.

CHART 2
CHART 2

SLOVENIA: LABOR MARKET DEVELOPMENTS

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by the Slovenian authorities.

Faced with this situation on the labor market, the Slovenian authorities’ strategy has been multifaceted. First, in order to foster labor market flexibility, steps have been taken to reduce firing costs. Second, active labor policy measures have been adopted (e.g., specific employment subsidies to reduce labor cost, retraining). Third, a well-developed safety net has been maintained, including relatively generous unemployment benefits and easy access to early retirement for older workers. Fourth, and more recently, wage agreements since May 1994 have been aimed at containing wage inflation to increase labor demand.

a. Employment

Total employment has fallen sharply since 1989. Between 1990 and 1992, it fell by 13.9 percent and, subsequently declined by another 4 percent over 1993–1994. This dramatic change was accompanied by a reallocation of labor from socially-owned to private enterprises (large and small). However, sectoral reallocation within the enterprise sector was more limited (Table 33).

Over 1990–94, employment in large firms dwindled. In particular, due to restructuring and privatization, socially-owned large firms reduced their labor force by almost 45 percent. By the same token, the share of mixed and privately owned firms surged from about 5 percent to more than 25 percent of large firms’ employment (Table 36). Over the same period, the share of small firms and self-employed in total employment rose strongly from 14 percent in 1990 to 19 ½ percent in 1994.

These numbers, however, may well overestimate the undergoing transformation. Anecdotal evidence suggests that some of these changes were merely a way to circumvent labor legislation or to minimize the tax burden. For instance, many social enterprises transferred their workers and assets to private subsidiaries as a means to avoid workers’ councils veto on labor shedding. The mother company therefore became a financial holding company with a limited number of employees. Also, the growing number of the self-employed may be due to a more advantageous fiscal status (see Chapter IV on Fiscal Developments).

Sectoral reallocation has been somewhat limited. The staffing of a new administration has increased employment in the public sector by 6 ½ percent between 1990 and 1994. Within the enterprise sector, changes have been much less marked. Overall, the share of services has edged up slightly, in particular for financial services, but remains distinctly lower than the EU average (Table 2).

Table 2.

Employment by Sectors of Activity in Slovenia and the EU

(In percent)

article image
Source: Labor Force Survey—Statistical Office of Slovenia (1994).

The unskilled labor has borne the brunt of the labor force adjustment. In 1988, for the labor force as a whole, the average number of schooling years was 9.3, close to the average for socialist economies (Vodopivec and Hribar-Milic, 1993). Since then, the proportion of highly educated people has increased. Higher education has proved to be a useful shelter against unemployment risks, and the return on education has increased. The impact of education on unemployment (Table 3), would be even more visible if these numbers are corrected for the fact that early retirement has reduced the number of unemployed older workers who, on average, have a lower number of years of education.

Table 3.

Structure of Labor Force and Unemployment by Level of School Attainment, May 1994

(In percent)

article image
Source: Labor Force Survey—Statistical Office of the Republic of Slovenia (SORS)

Efforts were made even in the former SFRY to foster employment flexibility by reducing firing costs. Legislation enacted in 1988–89 gave employers the right to lay-off workers. Redundancy payments, however, remained substantial. In particular, the advance notification period for lay-offs was 24 months and workers were not expected to work during that spell, often dubbed as “paid vacations”.

The Slovenian authorities have facilitated labor shedding by reducing administrative hurdles and further lowering firing costs. In 1991, amendments to the Labor Code reduced the advance notice period to 6 months for “technologically redundant” workers. In the case of mass lay-offs, however, workers” councils have retained a significant say and firms have to prepare restructuring programs for redundant workers, co-financed by the state which reimburses up to 50 percent of the associated costs. In 1991, the government co-financed 21,223 redundant workers. A large majority were put on “paid vacation”, 2,499 received a lump-sum payment, 1,082 chose early retirement and 172 were trained (Abraham and Vodopivec (1993)).

Early retirees were offered the possibility to “buy-back” qualifying years to increase their pension benefits. Given the generous conditions offered, these measures attracted relatively large flows of new pensioners: 6,600 in 1991 and 4,600 in 1992. However, it became increasingly clear that early retirement was creating a substantial threat for long-term financial balance of the Pension Fund. Therefore, mass lay-off plans have reduced their reliance on early retirement and the influx of early retirees has ebbed in 1994 and 1995 (see Appendix II).

b. Unemployment and labor market policies

In the mid-1980s, the unemployment rate fluctuated around 1 ½ percent; it rose sharply after the beginning of the transition and peaked in December 1993 at 15 ½ percent. Since then, the unemployment rate has recorded a slow decline to 13.4 percent by May 1995. There is, however, a puzzling divergence between registered unemployment rates and estimates resulting from the Labor Force Survey. 1/ Divergences between the two indicators are not uncommon in other countries. However, for Slovenia, the difference is the most sizeable. 2/

The adjustment in labor force participation has been key to contain unemployment. Between 1990 and 1994, about 150,000 jobs were lost; the number of unemployed, however, only increased by 83,000, with the difference resulting from lower labor force participation, in particular for older workers (Table 37).

Table 4.

Unemployment Rates in Eastern and Central Europe

(1993 2nd quarter—In percent)

article image
Source: Koltay (1994) and Labor Force Survey—SORS (1994).

The exit rate out of unemployment is fairly low. A study of displaced workers by Orazem et al. (1995) finds that half of the displaced workers in January 1989 were still unemployed by August 1992. As a result, the proportion of the long-term unemployed (more than one year) reached 48 percent in May 1994.

Unemployment benefits are relatively generous although eligibility requirements have been progressively tightened since 1991. For the first three months, benefits represent 70 percent of the average wage during the three months preceding the unemployment spell. Thereafter, the replacement rate diminishes to 60 percent. Depending upon past employment history, the duration of entitlement ranges from 3 months (for past employment superior to 9 months) to 2 years (greater than 20 years). Benefits are normally payable on a monthly basis; workers receive a lump-sum payment if it can be used to obtain employment.

To complement unemployment insurance, assistance is provided to job seekers who are no longer eligible for unemployment benefits on a means-tested basis—income per family member below 80 percent of the minimum wage. Job seekers are also entitled to the reimbursement of moving expenses (to foster geographical mobility), to health insurance, and to certain in-kind benefits (e.g., free meals).

Some observers have argued that unemployment benefits are too generous, compared for instance to OECD countries. In particular, the exit rate out of unemployment rises sharply immediately before job-seekers become non-eligible for benefits. One may note, however, that the budgetary costs of unemployment benefits have been contained below 1 ½ percent of GDP, on average over 1991–94.

A key feature of labor market policies in Slovenia is the importance of active policies. Training programs, in particular, play an important role for both workers and the unemployed. The former can receive vocational on-the-job training, typically for a period of 4 to 5 months. A system of fixed-term contracts also attempts to help first-time job seekers. The unemployed receive 80 percent of the minimum wage if they enroll in a training program.

A substantial employment subsidy program aims at preventing bankruptcies and unemployment. The number of displaced workers due to bankruptcy procedures had increased drastically between 1989 and 1991, from 1,600 to 19,200. In July 1991, the government suspended the initiation of bankruptcy procedures by the Agency for Payments, Supervision and Information (APPNI), formerly the Social Accounting Service (SDK), and to preserve employment subsidized financially distressed firms. Subsequently, the flow shrank to 6,700 in 1992 and 9,700 in 1993. To restore their economic viability, these distressed firms may receive various subsidies (e.g., discount on electricity bills or tax reliefs).

c. Wages

As in other transitional economies, real wages described a U-shaped pattern, falling in the initial stage of the transition and then recovering. The fall in wages had begun as early as 1989—the cumulative fall between 1989 and 1992 was 40 percent, with a large fraction of the decline taking place in 1990. Like in other transitional economies, however, in-kind benefits provided by firms smoothed the adjustment. Over 1992–94, the recovery was as brisk as the decline itself: real net wages bottomed out in February 1992, and by end-1992, had regained a large fraction of the initial decline. Ultimately, real wages in 1994 were at the same level as in 1990 (Table 34).

The strong growth in real wages between 1992 and 1994 caused two main concerns. First, there was a fear that it would fuel inflationary pressures. Second, to contain unit labor costs firms would have to achieve strong productivity gains (Table 38). Such a high-wage/high-productivity strategy seemed undesirable in a high-unemployment context.

Collective bargaining has been the main vehicle to check wage growth. In 1993, no agreement was signed and wage inflation reached 16 ½ percent. The subsequent inflationary pressures convinced trade-unions and the Chamber of Commerce—which represents the enterprise sector—that a wage agreement was required. A March 1994 Agreement, that had been reached between the two parties, was transformed into a law by Parliament in May. Its main disposition was that nominal wage growth would be below the rate of inflation for the rest of 1994. As an incentive to comply with the Agreement, the government imposed an additional wage tax which could reach up to 100 percent of the excess wage.

The Agreement moderated wage growth to 6 percent in 1994. The results were, however, not completely satisfactory for two reasons. First, certain clauses permitted firms to grant wages exceeding the agreed-upon levels without penalty. A number of firms elected to do so; some others were willing to pay the excess-wage tax. Second, the wage agreement had a limited horizon which created the possibility that wage growth could accelerate upon the termination of the Agreement at end-year.

The Wage Agreement for 1995—which covers the period April 1995 to March 31, 1996—was signed in mid-May 1995. This Agreement has more far-reaching goals. In addition to assisting in cooling down inflation, the Agreement states that wage moderation is key to fostering employment growth. Therefore, the parties—the government, employers and employees—agreed that wages be only imperfectly indexed to prices over 1995. Given the spurt in wage growth during the “lame-duck” period between the 1994 and 1995 Agreements, the implementation of the agreement requires a reduction in nominal wages in July 1995. This reduction is to be enforced by the APPNI. The wage Agreement is also intended to be part of a broader agreement on macroeconomic policies whereby the government commits itself to deliver appropriate fiscal and monetary policy to stimulate employment growth.

The 1995 Agreement will, however, do little to alter the wage structure in Slovenia. The wage structure was compressed by workers and government interventions in the former SFRY. Under self-management practices in the former Yugoslavia, workers’ councils had a tight grasp on relative wages at the firm level. Also the government’s objective was to level down wage differentials amongst firms through the use of discretionary taxes and subsidies. For example, loss-making firms were exempt from tax payments and received subsidies.

There is some evidence that wage dispersion has increased in Slovenia, in particular, to provide a higher return to education (see Orazem and Vodopivec (1993)). Amongst sectors, however, this wage differentiation has remained limited. To gauge the magnitude of these trends, the standard deviation of average sectoral level was calculated for single-digit industries. The calculations indicate that wage dispersion increased substantially before 1991. This differentiation process subsequently came to a halt. Similar trends are being observed in other transitional economies (W. Perraudln and T. Pujol, 1995). It may be conjectured that centralized wage bargaining in an inflationary environment, where workers try to achieve perfect indexation, is unlikely to deliver a large increase in sectoral wage differentiation.

Table 5.

Standard Deviation of Sectoral Wages

(In percent)

article image
Source: Labor Force Survey—SORS (1994).

II. External Sector Developments

1. Background

Slovenia has a highly open economy with the ratio of external trade to GDP that averaged over 120 percent during 1992–94. Even as a part of the former SFRY, its geographical proximity to Western markets made Slovenia a springboard for industrial production and processing for these markets; with only 8 percent of the population of the former Yugoslavia, it contributed almost 30 percent of exports. The close contacts, and the existence of a sizeable market presence, conferred Slovenia with advantages that have stood it in good stead once it gained independence. Slovenia adapted its trade patterns quickly to the loss of markets in the former SFRY and the CMEA. 1/ In addition, through a number of free trade agreements with the EU and countries of Eastern and Central Europe, Slovenia has gained better access to export markets.

Estimates have been made of trade losses that Slovenia experienced from the disintegration of the former SFRY and the CMEA. 2/ Besides losses to merchandise and processing trade, uncertainties arising from the security situation caused a precipitous decline in tourism earnings and cross-country transportation, both major sources of foreign exchange earnings for the economy.

Principal developments in the external accounts over the period 1992 to the first half of 1995 relate to: (i) a persistent surplus in the current account; (ii) recovery in, and reorientation of exports from the former Yugoslavia to Western markets; (iii) recession in partner countries and import boom of consumer goods in 1993; (iv) steady rebound in tourism and transportation earnings; (v) sizeable private capital inflows from the second quarter of 1993 to the fourth quarter of 1994; and (vi) substantial accumulation of international reserves over the entire period. Following a brief description of trends, their proximate determinants are identified and analyzed.

2. Balance of payments

The balance of payments for Slovenia has been characterized by current account surpluses of varying magnitudes and large net capital inflows in 1993–94 (Table 41). Together these flows have contributed to a rapid accumulation of net foreign assets by the banking system (Chart 3). Total net international reserves of the banking system reached US$3.5 billion at end-July, 1995 rising from a meager US$0.4 billion at end-1991. Net international reserves of the BOS increased from a paltry US$112 million at end-1991 to US$1.7 billion at end-July, 1995, while the foreign assets of deposit money banks (DMBs) expanded from about US$300 million to US$1.5 billion. This sharp increase in net foreign assets over the period end-1991 to May, 1995 was financed primarily by cumulative current account surpluses (Table 6).

CHART 3
CHART 3

SLOVENIA: FOREIGN EXCHANGE HOLDINGS OF THE BANKING SYSTEM

(in millions of US dollars)

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by the Slovenian authorities.
Table 6.

Contributions to Reserve Accumulation

article image
Sources: Bank of Slovenia; and staff estimates.

a. Current account

The current account has been consistently in surplus during 1992–94 but these surpluses have varied substantially during this period (Chart 4). The current account surplus rose to 7 ½ percent of GDP in 1992 from 1 percent of GDP in 1991, only to recede to 1 ¼ percent in 1993 before rising again to about 3 ½ percent in 1994. Except for 1992, the trade account was in deficit. Thus, the current account surplus stemmed from large surpluses on the net services account related to tourism earnings.

CHART 4
CHART 4

SLOVENIA: DEVELOPMENTS IN THE CURRENT ACCOUNT

(in millions of US dollars)

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by the Slovenian authorities

The variations in the current account surplus cannot be explained by movements in the fiscal position. The fiscal accounts were in modest surplus (about ½ percent of GDP) in 1992–93 and slipped into a small deficit in 1994. This swing of about 1 percentage point of GDP compares to changes in the current account surplus of 6 percent of GDP. Thus, the nongovernment sector has been primarily responsible for the large current account surplus and variations in it.

The saving-investment balance offers a useful perspective for analyzing the level of, and movements in, Slovenia’s current account surplus. In 1992, the investment/GDP ratio was low owing to uncertainties related to the future course of economic policies in the newly independent country. The privatization process, in particular, was still the subject of an unresolved debate. In each subsequent year, the investment/GDP ratio has risen. Indeed it has increased by 4 ½ percentage points of GDP from 1992 to 1994. Thus, the current account surpluses can be attributed to domestic savings ratios of the private sector. Domestic savings were nearly 27 percent of GDP in 1992 and fell to 21 ¼ percent in 1993, as Slovenia experienced a consumption boom that sucked in imports, particularly of consumer durables. This drop in the savings ratio of nearly 5 percentage points of GDP explains 80 percent of the decline in the current account surplus.

(1) Trade balance

Over the period 1992–94, Slovenia had a deficit in the trade balance with the exception of 1992. The swing in the trade balance of about US$1 billion between 1991 and 1992 is fully accounted for by the redirection of exports to Germany and France following the loss of markets in the former SFRY and the CMEA. Moreover, 1992 was the first full year of independence and, for the first time, trade with republics of the former SFRY was recorded as external trade. Slovenia had a surplus, amounting to almost US$300 million, on this account. The large surplus on the trade account US$790 million (6 ½ percent of GDP) in 1992 was completely wiped out in 1993. The swing in the trade balance by almost 7 ¾ percent of GDP led to a deficit of about US$150 million (1 ¼ percent of GDP). The shift was caused by the decline in export earnings by US$600 million, due to the recession in Western markets. At the same time, a rapid expansion in domestic household incomes generated a boom in consumer goods imports, which increased by US$400 million. In 1994, the trade balance remained virtually unchanged in dollar terms as merchandise exports and imports both grew by 11 ½ percent. The recovery in Western markets was the primary cause of the expansion in exports. The growth in imports, in contrast to 1993, was focused entirely in capital and intermediate goods; consumer goods imports were almost unchanged in dollar terms.

(a) Merchandise exports

The high degree of openness of the economy and the geographic concentration of trade makes the fortunes of Slovenian exports highly susceptible to the level of economic activity in the EU; econometric estimates of the dependence of Slovenian exports earnings place the elasticity of exports earnings to import demand in four principal European trading partner countries (with a lag of one quarter) at 0.61. 1/ It is plausible that, in the period following independence, as exports have become more focused upon countries in the EU, this elasticity may have risen. The shortness of the time series precludes a firm analysis, but preliminary evidence using quarterly data for the period 1992 to the first quarter of 1995 indicates that this elasticity may be greater than unity.

In addition, exports earnings in U.S. dollar terms are also sensitive to movements of the U.S. dollar against the German mark and the Italian lira. Almost 60–70 percent of Slovenia’s exports are denominated in these two currencies. A decomposition of export values into price and quantities is presented in Table 7.

Table 7.

Decomposition of U.S. Dollar Export Values

article image
Sources: Data provided by Bank of Slovenia; and staff calculations.

Following the collapse of the internal market in the former SFRY, Slovenian firms were quick to expand sales to Western markets (Table 42). Merchandise exports to non-Yugoslav markets grew by 8 percent to US$4.2 billion in 1992, and to the EU by 14 percent. Total exports, including the former republics, stood at US$6.7 billion in 1992. The recession in Europe was the primary cause of a 9 percent decline in export earnings in 1993; earnings from exports to the EU fell by 5 percent. Exports to markets in the former SFRY fell by another US$545 million to US$1 billion in 1993; the only factor moderating the fall in export earnings was a reorientation to countries of the former CMEA. The recovery in Europe in 1994, coupled with a weakening of the dollar against the deutsche mark, generated a 12 percent increase in returning export values to US$6.8 billion in 1994.

The volume of export fell by 3 ½ percent in 1993 and recovered by about 5 percent in 1994. An examination of the trends, by quarter, reveals that while the decline in exports continued late into 1993, export volumes had already begun a decisively upward trend as early as the first quarter of 1993. Export volumes demonstrated a rising trend over the remainder of 1993 and throughout 1994; the index of export volumes was 14 percent lower in the first quarter of 1993 compared to the same period in 1992, but had recovered almost all the lost ground by the first quarter of 1995.

The substantial change in the geographical destination of exports is evident in the halving of the share of exports to the former SFRY and the CMEA in total exports to 15 percent and a corresponding increase in the share of the EU from 61 to 66 percent over 1992–94. Export activity within the EU is also concentrated, with four major trading partners—Germany, Italy, France and Austria—accounting for almost 90 percent of exports to the EU (and 55 percent of all exports). Croatia is a large and important trading partner accounting for 10–12 percent of total exports.

The commodity composition of Slovenian exports continues to be dominated by manufactured products which account for over 80 percent of total exports (Chart 5 and Table 43). Within manufacturing exports, however, the relative contribution of commodity groups (classified according to SITC codes) has changed in response to the differential sectoral effect of the need to shift to more competitive export markets. Sectors also differ in the degree to which they are affected by changes in exchange rates. The loss of traditional markets was most acutely felt by machinery and transport equipment sector whose share in total exports fell by 10 percent to about 30 percent in 1992. With the relative share of all other commodity groups almost constant, the share of other manufactured goods in total exports (in particular textiles, furniture and footwear) rose even at virtually constant export values.

CHART 5
CHART 5

SLOVENIA: COMPOSITION OF TRADE

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by Slovenian authorities
(b) Merchandise imports

Trends in merchandise imports have been related to developments in real wages, the level of external competitiveness and the pace of investment activity. While imports were initially driven almost entirely by consumption goods, capital goods import increasingly became the principal mover after the second quarter of 1993 as investment expenditures rose. The level of external competitiveness (as measured by a CPI-based real effective exchange rate) remained virtually unchanged from 1992 to the second quarter of 1994 and, therefore, had little impact on import volumes. There is preliminary evidence that the appreciation of the real exchange rate that began in the third quarter of 1994 has generated a shift of domestic demand to imported goods.

Merchandise imports were depressed in 1991 and 1992. Uncertainties caused by the transformation process and low levels of economic activity generated a decline in imports of capital goods; slow export growth (given the high degree of processing activity) led to lower imports of intermediate goods; and the fall in real incomes that had taken place in the preceding years induced a squeeze on consumption goods imports. 1/ The resumption of real wage growth in 1992 reversed this trend and consumer imports spurted sharply in the first quarter of 1993. The boom in consumer durables was especially marked, with imports of road vehicles, which had averaged US$475 million over 1990–92, rising to over US$755 million in 1993. Imports of consumer goods moderated markedly in 1994. However, total import volumes rose even further as the recovery in industrial production in mid-1993 gave an impetus to capital-goods imports. In dollar terms, imports of capital goods rose by 11 percent in 1993 and by a substantial 22 percent in 1994. Imports of intermediate goods also recovered in 1994 increasing by 13 percent reflecting the importance of processing and re-export activity and its close relationship to export growth.

The geographical composition of imports—like that of exports—showed a continued move toward Western Europe during 1992–94 (Table 45). The share of the EC and EFTA rose from 60 percent to almost 70 percent. During the same period, the share of the former SFRY republics and CMEA dropped from 19 percent to 13 percent. The commodity composition (Table 46) mirrors the trends in imports by end-use over the period. In general, about 60 percent of imports consist of manufactured goods, of which more than half is machinery and transport equipment. Raw materials and fuels account for some 17 percent, chemicals slightly more than 15 percent, and food and live animals around 5 percent.

(2) Services

The underlying strength of the current account largely reflects the traditional strong surplus in the service account (Table 46 and Chart 6). The surplus averaged US$920 million or 7 ⅓ percent of GDP in 1988–90, but declined to 3 percent in 1991. 2/ The surplus on the services account has recovered, albeit slowly, to over US$600 million (4 percent of GDP).

CHART 6
CHART 6

SLOVENIA: NON-FACTOR SERVICES CURRENT ACCOUNT

(in millions of US dollars)

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by Slovenian authorities.

Principal contributors to the surplus on the services account are tourism earnings and receipts from cross-country transportation. Tourism and transport sectors registered a drastic decline in 1991, as a result of the breakup of the former SFRY and the deterioration in the security situation in the region. Total receipts fell by almost US$680 million to 8 percent of GDP. The situation improved greatly in 1992, when tourism receipts almost doubled to US$670 million and processing fees increased by 18 percent to US$180 million. These favorable developments more than offset a further 20 percent decline in transport receipts and total receipts from non-factor services rebounded to US$1.2 billion or 11 ½ percent of GDP in 1992. The recovery in both sectors has been steady over 1993–94. Some of the increase in receipts, especially in 1994, is due to valuation changes caused by the depreciation of the U.S. dollar against European currencies. 1/ The quantity magnitudes, however, reflect the rebound more clearly. Registered night stays by foreign tourists in Slovenia, which had peaked at almost 4,000 thousand in 1989, fell precipitously to 970 thousand in 1991. The recovery began in 1992 and registered stays recovered to around 2,500 in 1994. Non-factor expenditures, which consist mainly of transport and travel expenditures, have averaged 3–5 percent of GDP in recent years.

Receipts and expenditures for factor services account for a small part of the current account balance (1–2 percent) contributing an overall deficit to the total. With relatively small direct and portfolio investment into and out of Slovenia, interest receipts and payments constitute over 95 percent of the investment income account. Due to measurement and recording problems, payments on account of labor incomes to non-residents (in particular, to migrant labor from Croatia) may be underestimated.

Private transfers and labor remittances fluctuated between US$70–100 million per year until 1991, when they virtually dried up as a result of the uncertain environment in Slovenia. The recovery was swift and nearly complete with inflows averaging US$100 million in 1993–94. Official transfers, consisting mainly of technical assistance, averaged a little over US$60 million over 1992–94.

b. Capital account

(1) Trends in and composition of capital flows

Reflecting similar trends in other economies in transition in Central and Eastern Europe, a net outflow of capital from Slovenia occurred in 1990 and 1991. 1/ 2/ In the case of Slovenia, these outflows were almost entirely a reflection of the uncertainties associated with the break-up of the former SFRY. Net capital inflows resumed in 1992 (even though net errors and omissions showed a large outflow) and became substantial in 1993 and 1994 averaging US$480 million or 3 ¾ percent of GDP (Chart 7). Compared to other transitional economies, however, the composition, nature and the tinder underlying factors for the flows are somewhat different. In particular, these flows were predominantly private in nature. A breakdown of these components is presented in the following table:

CHART 7
CHART 7

SLOVENIA: DEVELOPMENTS IN THE CAPITAL ACCOUNT

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source Data provided by Slovenian authorities.1/ Includes net errors and omissions.2/ Defined as the nominal interest rate on 31–90 day time deposits minus the German interest rate on 3-month time deposits less than 1 million marks minus the annualized exchange rate crawl (to proxy for expectations of movements in the exchange rate).3/ Defined as household currency and deposits and loans of enterprises.4/ Defined as the difference between the nominal interest rite on foreign exchange deposits in Slovenia and the interest rate on German 3 month time deposits less than 1 million marks.
Table 8.

Components of the Capital Account

article image
Sources: Bank of Slovenia; and staff estimates.

In a number of the other transitional economies, the net outflows of the late 1980s and the early 1990s were associated with large debt service payments and difficulties in obtaining new financing due to political uncertainties. Similar factors were at work in the former SFRY; however, the introduction of “internal convertibility” under which citizens could purchase foreign exchange freely led to a large exodus of households’s foreign currency holdings abroad in 1990 and 1991. The recorded capital account (excluding net errors and omissions) for Slovenia displayed a net outflow of a little over US$150 million in 1991. Direct foreign investment all but dried up. Outflow of currency and deposits by households contributed two-thirds to the total net outflow of US$150 million. 1/ This substantial accumulation of currency and deposits was ripe for a return as soon as domestic political and economic conditions stabilized.

Following a return to political and monetary stability in 1991, the capital account (excluding errors and omissions) turned around in 1992. Net capital inflows were roughly equivalent to the outflows in 1991. Direct investment tripled to US$115 million, albeit from the low level of 1991, and together with loans of enterprises (net of repayments amounting to US$60 million), accounted for all of the surplus of US$175 million (1 ½ percent of GDP). The privatization of socially-owned houses and apartments induced a return of about US$120 million on account of household currency and deposits in the first four months; an outflow of an almost equal amount in the remainder of the year, due to varying and often negative uncovered interest differentials, left the contribution of this source negligible for the year as a whole. Large errors and omissions (US$280 million or 2 ¼ percent of GDP) preclude a more useful categorization of the transactions in the balance of payments in 1992. The contribution of the capital account to reserve accumulation in 1992 was small.

The capital account increased sharply in 1993 to US$425 million or 3 ⅓ percent of GDP. An important source of the change was the regained access to commercial credits by importers which contributed over US$100 million to the surplus. Inflows of currency and deposits by households and short-term loans by enterprises (about US$65 million each) were also important contributors to the inflow; loans by the official sector added another US$90 million. Direct foreign investment remained unchanged. The turnaround in inflows on account of households came around the middle of 1993, also the peak season for tourism in Slovenia. 2/

The capital account surplus grew even larger in 1994 to US$535 million or 3 ¾ percent of GDP. Inflows of currency and deposits grew to a veritable deluge in the first semester; net inflows amounted to US$375 million for the year as a whole, while loans by the enterprise sector (classified largely as “long-term”) more than doubled to US$225 million. Pre-payments on imports moderated the surplus by about US$150 million and may have been a response to expectations of a nominal appreciation set in motion by the slowdown in the rate of crawl of the exchange rate in mid-1994.

(2) Capital flows: An empirical analysis

The preceding account provides an indication of the factors that could help explain trends in the capital account. Econometric analysis of possible determinants of the capital account is, of course, subject to several caveats: (i) the balance of payments data is subject to statistical misclassifications; (ii) the available time series are of short length and high frequency; and (iii) the underlying structural relationships could still be in flux. Nonetheless, a preliminary analysis may shed some light, providing a more rigorous underpinning to policy judgements that need to be made in any case, and a foundation for further analysis.

Various measures of the capital account were used: (i) the capital account, including errors and omissions (KA); and (ii) the capital account, including errors and omissions but excluding loans by the official sector (KA1). Two narrower measures were also examined, focusing on private capital inflows: (i) Household currency and deposits and loans of enterprises (PKF); and (ii) Household currency and deposits only (HHCD).

With regard to explanatory variables, it is evident that movements in the capital account would be governed by secular changes in the economic (and political environment) and, given the high degree of openness of the capital account, by short-term variation in interest rates. The role that the return to political and economic normalcy may have played in the resumption of the inflows is evident in the early rise in the level of direct foreign investment which is typically governed by long-run considerations. Similar factors affected commercial credits received by importers though expectations about future movement in the exchange rate may also have played a role. It is to be expected that capital inflows are positively correlated with interest rate differentials, and the level of economic activity (increased investment expenditures would be financed by loans abroad and consumption expenditures may be financed by inflows of currency and deposits).

Two measures of interest rate differentials were used: (i) uncovered tolar interest rate differentials, defined as the nominal interbank rate in Slovenia minus the interbank rate in Germany minus the annualized rate of exchange rate crawl (to proxy for expectations of movements in the exchange rate); and (ii) interest differentials between domestic foreign currency and German time deposits. 1/ The level of economic activity was proxied by indices of industrial production (both total and capital goods). Rising activity, particularly investment, is expected to increase foreign borrowing or import financing. The large magnitude of the errors and omissions in 1991 and 1992 and the fact that the classification of foreign exchange flows between the current and capital transactions may not be completely reliable suggests that any empirical estimation of the determinants of the capital account as a whole would have to grapple with the issue of measurement error; the current account and tourism receipts were used to proxy, inter alia, for measurement errors, with a negative expected correlation.

OLS regressions were estimated using monthly data with contemporaneous and (a maximum of two period) lagged values of the regressors for the period January 1992 to April 1995. Equations were also estimated for the period January 1993 to April 1995 when the contribution of net errors and omissions was relatively small. Non-significant variables were sequentially dropped to obtain a parsimonious representation. The estimated equations are reported below:

article image

Equation I: 1/

KA = - 182.4 ( - 1.12 ) - 0.11 ( 0.73 ) IDTD + 0.03 ( 0.37 ) IDTD(-1) + 0.08 ( 1.16 ) IDTD(-2)- 0 .87 (-4 .80)** CA - 0 .90 (-0 .83) IGP + 1 .95 (1 .63)* IGP(-1) + 1 .42 (1 .18) IGP(-2) Time period: January, 1992 to April, 1995 . Number of observations: 38 R 2 = 0 .59 Adjusted R 2 = 0 .49 F statistic = 6 .08 DurbinWatson: 2 .02

Equation II:

KA1 = - 113 .2 (-1 .23) + 0 .03 (0 .56) IDTD(-2) - 0 .79 (-5 .57)** CA + 1 .73 (1 .85)* IGP(-2) Time period: January, 1992 to April, 1995 . Number of observations: 38 R 2 = 0 .54 Adjusted R 2 = 0 .50 F statistic = 13 .15 DurbinWatson: 1 .99

Equation III:

PKF = - 83 .2 (-1 .45) + 0 .78 (3 .05)** IDTD + 1 .16 (1 .94)* IGP (-2) Time period: March, 1993 to April, 1995 . Number of observations: 26 R 2 = 0 .45 Adjusted R 2 = 0 .41 F statistic = 9 .58 DurbinWatson: 1 .92

Equation IV:

HHCD = 14 .86 (3 .70)** + 0 .93 (4 .07)** IDTD Time period: March, 1993 to April, 1995 . Number of observations: 26 R 2 = 0 .41 Adjusted R 2 = 0 .38 F statistic = 16 .59 DurbinWatson: 2 .02

The estimated equations, despite the data caveats, provide a reasonably good fit to the data. For the period as a whole, the current account and lagged industrial production are statistically significant determinants of the broad measures of the capital account, overshadowing interest differentials, suggesting that measurement problems may dominate the analysis at this high aggregated level. 1/ In general, the narrower measures of the flows do not provide as good a fit to the data as the broader measures for the longer sample period. However, contemporaneous values of both measures of interest rate differentials persistently appear as statistically significant. 2/

The goodness-of-fit and the diagnostics of the estimated equations improve substantially when private capital flows are the dependent variables and the time period is truncated to March 1993 to April 1995. 3/ In addition to interest rate differentials, which remain highly significant, lagged values of industrial production also become significant explanatory variables for PFK. Almost all variables, with the sole exception of interest rate differentials, drop out when HHCD is used as the dependent variable suggesting that interest differentials may have been the primary factor behind these inflows.

3. External dept

Slovenia has made considerable headway in regularizing its creditor relations. The apportionment of debt of the former SFRY among its successor states continues to be an ongoing process and involves a complicated set of issues. The determination of Slovenia’s share has employed a number of principles. According to BOS data, using a final beneficiary principle, Slovenia’s total medium- and long-term debt assumed from the former SFRY was US$1,765 million at end-1991. 1/ By end-1994, medium- and long-term debt amounted to US$1,659 million, of which about 30 percent was owed to multilateral institutions (excluding the Fund), 20 percent to Paris Club creditors, and 40 percent to commercial banks (Table 47). 2/ Fund credit to the former SFRY was divided among the successor states in accordance with the Fund quota key with Slovenia’s share at 16.39 percent. The same principle was used by the World Bank to determine Slovenia’s share of the debt of the former SFRY that could not be allocated according to the final beneficiary principle.

a. Official creditors

With regard to bilateral official creditors, initial contacts between the Slovenian authorities and the Paris Club secretariat commenced in June 1992. 3/ These contacts—conducted through an exchange of letters between the MOF and the secretariat—sought to determine the principles for allocating Slovenia’s share of the official bilateral debt of the SFRY. No Agreed Minute was signed by the creditor governments and Slovenia, as is customary under a Paris Club rescheduling; also there was no reconciliation of the amounts owed to individual official creditors.

The exchange of letters, inter alia, established principles to be observed in bilateral negotiations between the Republic of Slovenia and individual Paris Club creditors. These principles were: (i) Slovenia would assume 16.39 percent of the unallocated debt of the former Yugoslavia; (ii) the terms and conditions of the original rescheduling agreements would continue to apply; (iii) repayments on existing late payments on the consolidated debt would be staggered over a period of three years; and (iv) creditors had no objection to the use of the final beneficiary principle, where practicable. However, in defining Slovenia’s share, debt owed or guaranteed by entities located on Slovenian territory should also be taken into account; and (v) internally transferred obligations would be settled on a bilateral basis with the caveat that, if they were effected without the prior consent of the creditor, the substitution of the debtor would be invalid.

Point (iv) has been a source of problems in reconciling amounts owed and completing bilateral negotiations. This point establishes two criteria for allocation: (i) a final beneficiary criterion; or (ii) a territorial criterion. The former criterion is determined by the economic benefit derived from the loan or associated project rather than the territory wherein the bank borrower/guarantor might be located. Implementation of the final beneficiary principle was considered possible because all borrowings by the former Yugoslavia were registered with the national bank of each republic or autonomous provinces and records were maintained by domestic commercial banks of borrowings and repayments made by final beneficiaries. However, over the various reschedulings of the original official debt by Paris Club creditors, a complicated and varied set of relationships arose between the National Bank of Yugoslavia (NBY), domestic commercial banks and creditors. The NBY and domestic commercial banks were variously named as the guarantor, agent and borrower. In addition, interlocking relationships arose between commercial banks and final beneficiaries in various republics and have been a source of dispute. Lack of agreement between Slovenia and some creditors on the primacy of these two criteria led in some cases to a prolonged dispute over the amount of Slovenia’s debt.

Bilateral negotiations began at different dates with official bilateral creditors and are in various phases of completion. Initial negotiations with Belgium, France, Germany and the Netherlands have been completed but require review by a parliamentary committee, signature and, finally, ratification by the Slovenian parliament.

b. Commercial creditors

Commercial bank debt of the former SFRY was restructured under the New Financing Agreement (NFA) in 1988 and, as of June 2, 1995 amounted to US$4.6 billion. Based on the final beneficiary criterion, Slovenia’s obligations were estimated at US$377 million. Under the terms of the NFA, commercial banks in the various republics were named as debtors and bound by a “joint and several” liability clause, with the National Bank of Yugoslavia as the guarantor. Thus, each individual bank could be held liable by creditors for the entire amount owed under the NFA. Owing to the complications, inter alia, arising from the clause, Slovenia and commercial bank creditors were unable to reach agreement on Slovenia’s share of the debt. Pending the resolution of the issue, once principal repayments began to come due in January 1994, Slovenia limited debt service payments to debt owed by Slovenian final beneficiaries and on the unallocated debt (using the Fund quota key). These amounts were placed in a fiduciary account with Dresdner Bank, Luxembourg.

On June 8, 1995 an agreement in principle was reached between the International Coordinating Committee (ICC) representing creditors and Slovenia. Under the terms of this agreement Slovenia would assume 18 percent of the outstanding debt. 1/ The agreement has to be approved by two-thirds of the creditors. It will then have to be ratified by the Slovenian parliament. The actual amount of Slovenia’s debt would depend on the strength of the U.S. dollar between now and completion of the deal, and to a lesser extent on the ongoing process of debt reconciliation. This amount would be further reduced owing to the exclusion of “connected persons” from the transaction. Connected persons were defined on two lists prepared separately by the U.S. State Department—the OFAC list—in conjunction with enforcing sanctions on Serbia/Montenegro and by an arbitrator proposed by Slovenia and accepted by the ICC.

The agreement involves the issue of new Slovenian bonds, denominated in U.S. dollars and deutsche marks, in exchange for claims held by eligible creditors. Connected persons are not eligible for this exchange. The transaction is structured as a “qualified asset transaction” and no cash payment would take place. Thus, the sharing provisions of the NFA are avoided. The new bonds issued by Slovenia will consist of two series: (i) a “Base Series” in a principal amount equal to 18 percent of the maturities falling due under the NFA; and (ii) an “Additional Series” with the market value of the principal amount, upon issuance—Additional Series Market Value. (ASMV)—to equal 18 percent of the past due principal and interest under the NFA. The final maturity date (2006) for the new bonds would be the same as that for the NFA. Thus, the bonds would be repaid over 11 years—presuming the deal is closed towards the end of this year—instead of the original 13 years. All bonds are to bear interest at LIBOR plus 13/16 percent. Interest rate on the “Additional Series” will be reset when the first interest payment comes due, if required, at a spread over LIBOR (not to exceed LIBOR plus 1 ½ percent) to ensure that, at the reset date, these bonds will have a market value equal to the ASMV.

The agreement in principle stipulates that, with the consent of creditors holding more than two-thirds of Refinancing Loans, Slovenian entities would be released of the “joint and several” liability clause of the NFA. It also provides for Slovenia to “indemnify and hold harmless” the Agent and members of the ICC if they are subjected to legal expenses on account of the intended transactions in the agreement.

4. Trade system 2/

Slovenia maintains a relatively liberal exchange and trade regime. Pending the effectiveness of the new Law on Customs Tariffs, the tariff regime of the former SFRY’s remains in effect with tariff rates varying from 0 to 25 percent. Including import surcharges, weighted tariffs average 14 percent in 1994. Slovenia imposes quotas on imports of certain textiles and agricultural products. Import duties have been reduced temporarily by 50 percent for imports of raw materials and intermediate products, and by 80 percent for capital goods, provided the items imported are not produced domestically. In November 1992, the equalization tax on imports was reduced from 8 ½ percent to 1 percent. There is also a 1 percent customs clearance fee.

The new Law on Customs Tariffs will be effective January 1, 1996. Tariff reductions are contemplated in the new law. Protection for import-substituting industries will be practically unchanged from the current low levels as tariff rates will be low and will apply only to imports from countries with which Slovenia does not have a free-trade agreement. The new average tariff rate would be reduced to 10 ½ percent (compared to the existing rate of 12 ½ percent) based on 1994 import patterns. The tariff structure would also be rationalized. The equalization and customs clearance taxes will be abolished.

Slovenia became a full member of the GATT on October 30, 1994 and is implementing all obligations. It also signed the Marrakesh agreement and became a founding member of the World Trade Organization (WTO) on December 23, 1994; the agreement was ratified by Parliament in June 1995. Slovenia has implemented all tariff and other obligations, with the exception of those relating to agricultural products, stemming from WTO membership starting January 1, 1995. In the area of financial services, Slovenia will harmonize with the frameworks of the General Agreement of Trade and Services (GATS) and the EU.

The process of joining the EU began with exploratory discussions in December 1993 and negotiations for associate membership commenced on March 15, 1994. Following the completion of technical discussions, an association Agreement was initialed in June 1995. Harmonization of the rules of origin with the EU is under way; this would lead to some diminishing of competition from non-EU members.

With regard to the likely impact of obligations and privileges arising from membership in the various international forums, in particular on the balance of payments, preliminary analyses suggest that the impact on export performance is expected to be marginal; the effect on imports is expected to small initially, but rising over the medium-term. It is also expected that stiffer competition from large firms will lead to some sectors being dominated by larger firms.

In an effort to gain market access and reduce the geographical concentration of exports, Slovenia has entered into free trade agreements with a number of countries of Central and Eastern Europe including the Czech Republic, Slovak Republic, Hungary and Poland. Slovenia also has a free trade agreement with EFTA and intends to pursue the possibility of free trade agreements with the Baltic states (contacts have already been initiated with Latvia and Lithuania) and intends to join the CEFTA.

III. Monetary and Exchange Rate Developments

1. Overview

Slovenia has made excellent progress toward nominal stability since introducing the tolar at the time of monetary independence (October 1991). Double digit monthly inflation was reduced to around ½ percent by mid-1995 and the tolar exchange rate against the DM has stabilized (Chart 8). Real money aggregates fell initially but as confidence in the tolar was won, a strong sustained recovery in demand for money occurred, doubling real money balances during 1992–95 and beginning a “tolarization” of the economy (Table 49 and Chart 9). Slovenia’s approach to monetary policy was based on a monetary anchor with the money supply regulated via a reserve money program which targeted inflation through regulating growth of narrow money. Exchange rate policy shifted from an initial float, to a real rule from mid-1992 to mid-1994, and finally unannounced nominal stability.

CHART 8
CHART 8

SLOVENIA: INFLATION and EXCHANGE RATE

(Monthly Percentage Change)

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by the Slovenian authorities1/ Exchange rate in the bank-enterprise market.
CHART 9
CHART 9

SLOVENIA: REAL MONETARY AGGREGATES

(1992=100)

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by the Slovenian authorities

Monetary policy has gone through three major stages since monetary independence. 1/ In the initial stage (October 1991 through mid-1992), the BOS reduced reserve money to eliminate a large monetary overhang and to limit exchange rate depreciation. After losing three fourths of its initial value from October 1991 to January 1992, the tolar exchange rate stabilized during February-June 1992. The sharp fall in the real value of reserve and narrow money and the exchange rate would lay the foundation for lower inflation, a current account surplus, and a subsequent revival in real growth. Progress in the fight against inflation was achieved first. Inflation fell from 20 percent per month in October 1991 to about 2 ½ percent per month by mid-1992.

In the second stage (mid-1992 through mid-1994), BOS’s goals were to gradually reduce inflation and to protect external competitiveness and thereby orchestrate a revival in real output. A reserve money program was used to reduce inflation while a real exchange rate rule was implemented to protect competitiveness. Large sterilized purchases of international reserves and a complicated set of monetary regulations and policies were designed to induce banks to acquire international reserves. The nominal exchange rate depreciated approximately in line with inflation during this period, so that the real rule was realized (Chart 10). While inflation continued to fall during the second half of 1992, little progress was made in reducing inflation during 1993–94. Core inflation reached a plateau of about 1.1 percent per month while administered price inflation declined during 1993–94 accounting for the somewhat lower inflation in 1994 compared to 1993.

CHART 10
CHART 10

SLOVENIA: EXCHANGE RATE INDICATORS

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by the Slovenian authorities1/ A decline indicates a depreciation.

Sterilization became increasingly expensive in the face of swelling external surpluses and pressures for an appreciation. Furthermore, an incomes policy was adopted in May 1994 to offer another means to protect competitiveness. The third and current stage of monetary policy started in the last quarter of 1994 when the BOS reached the end of its willingness to sterilize. As a consequence the exchange rate stabilized against the DM (and appreciated against the U.S. dollar) and monthly inflation was halved to below ½ percent by mid-1995.

2. Establishing the new currency (1991/mid-1992)

In the first months after introducing the Tolar in October 1991, the BOS confronted difficult macroeconomic problems: double digit monthly inflation; a highly indexed economy (including broad money); no international reserves; low confidence in the new currency; and an absence of credibility in the BOS. The authorities’ main goal was to bring inflation down quickly from near hyper-inflationary levels. The BOS also wanted to acquire sufficient international reserves to facilitate trade, give itself more flexibility in monetary policy, and to enhance its credibility. The BOS managing board adopted a reserve money program to control inflation along with a floating exchange rate regime (Mencinger (1993), Bole (1995)).

From October 1991 to January 1992 the tolar lost three-fourths of its value, dropping from an initial level of SIT 13 per DM to SIT 50 per DM. This reflected the unfavorable macroeconomic conditions and the lack of confidence. The BOS changed reserve requirements, set-up controls over the central bank’s refinancing instruments, and subsequently reduced by 90 percent outstanding refinancing credits to banks. These polices led to a 30 percent nominal decline in reserve money and a 55 percent decline in real reserve money by January 1992.

In response, nominal variables began to stabilize in early 1992. The first sign was that the nominal exchange rate steadied at about SIT 51 per DM during February-June 1992. During February-April 1992, reserve money hovered at approximately its October 1991 level. Monthly inflation fell by two-thirds from its end-1991 level to 5 percent in April 1992. At this point, the BOS initiated its reserve money program that has remained in place ever since. Inflation was halved again to an average of 2 ½ percent per month during the second half of 1992.

Aided by an overshooting of the real exchange rate, a current account surplus of 7 ½ percent of GDP emerged during 1992 permitting Slovenia to substantially increase its foreign assets. The BOS acquired US$400 million during October 1991-June 1992, while commercial banks more than doubled their holdings of international reserves to nearly US$400 million by June 1992. In mid-1992 banks had to convert some of their foreign assets into tolars to obtain sufficient liquidity for their operations, owing to the BOS’s strict monetary policy.

3. Monetary developments in mid-1992 to 1994

During the second major phase of monetary policy, the primary goal remained the reduction in inflation, but the objective of promoting an economic recovery became more important (Bole (1995)). Large output declines that commenced in 1990 had continued into 1992, creating uncertainty as to when the economy would turn around. A more ambitious approach to controlling inflation based on an exchange rate anchor—now a viable option with international reserves 60 percent larger than base money—was rejected. It was feared that a nominal exchange rate anchor would lead to an excessive real currency appreciation, lower exports, and further depress output. Furthermore, the BOS was concerned that it did not have either the credibility or the requisite accompanying policies in place—specifically incomes policy—to support use of the exchange rate as a nominal anchor.

Reserve money programming was refined and became the primary means of controlling inflation. The reserve money programs was based on an unannounced inflation target, although it was widely known that the BOS was seeking to lower inflation. Reserve money growth would be slowed in the next month if inflation exceeded the described path. To further promote economic growth, the reserve money program was calibrated to accommodate increases in demand for money, particularly transaction demand (see Appendix I). Thus the monthly rate of increase in reserve money was increased with real growth in household income and real growth in transactions (when these data became available with a one to two month lag).

The other major aspect of monetary policy was an unannounced goal to depreciate the exchange rate to keep the real exchange rate constant (Bole (1994)). The BOS wanted the economy to retain the competitive edge that had resulted from the large currency depreciation in late 1991. The BOS also wanted support from a restrictive incomes policy to restrain real wage growth to assist in protecting external competitiveness—this was not forthcoming, however, until 1994. Additionally, fiscal surpluses during 1992–93 helped make room for sterilization.

The BOS had mixed results in achieving its inflation targets during 1993–94 as inflation appeared stuck in the range of 1–2 percent per month. End-year inflation was 23 percent in 1993 and was reduced to only 18 percent by end-1994 (compared to a budget target of 13 percent). The reduction in inflation that did occur in 1994 was entirely due to a decline in administered price increases; core inflation was 12 ½ percent in both 1993 and 1994. During 1994, administered price inflation was halved to 5 percent.

During 1993–94, Slovenia encountered entrenched inflation, reminiscent of other countries that used a gradual money-based approach to reduce inflation. Further decreases in inflation were stymied by the nominal depreciation of the exchange rate. The BOS was successful in preserving the real exchange rate for two years (from mid-1992 to mid-1994) through its sterilization intervention, but at a great cost. 1/

4. Monetary developments in 1994 to mid-1995

The BOS gradually initiated a major change in monetary and exchange rate policies during the second half of 1994. Several factors contributed to this change. Faced with substantial surpluses in the exchange market in 1994, the BOS intervened on a large scale to achieve its implicit exchange rate objective, while sterilization was undertaken to achieve its monetary target. Consequently both the BOS and banks began to acquire international reserves at an accelerated pace in 1994. 2/ The policy dilemma became increasingly frustrating and expensive as BOS profits began to plummet. 3/ During the second quarter of 1994, an incomes policy designed to limit wage increases to productivity gains was adopted by the Government and approved by Parliament. To some extent this introduced additional flexibility in the conduct of exchange rate policy. The need for a strict real exchange rate rule to maintain external competitiveness was lessened.

In mid-1994 the BOS developed and sold a new product—a warrant—in part to enhance its sterilization capabilities. 4/ With this instrument, the BOS was able to offset liquidity injections caused by foreign exchange purchases and repurchase operations (REPOs). 5/ However, when the bills became due in December 1994, the BOS only sterilized about two-thirds of the liquidity that was released (through REPOs). In addition, the BOS allowed banks—under the control of the Bank Rehabilitation Authority (BRA)—to sell foreign exchange directly to the BOS to decrease the level of outstanding liquidity loans to these banks. Finally, the BOS adopted a more passive approach to reserve money management by giving banks increased options to obtain tolar liquidity through direct sales of foreign exchange to the BOS. This series of activities amounted to unsterilized purchases of international reserves. Consequently, the BOS was regularly above its reserve money target from October 1994 to mid-1995. The growth of reserve money accelerated reversing the previous trend established during 1992–94. Reserve money growth, which had slowed to a 12-month rate of increase of 34.9 percent by end-September 1994, jumped to 56.9 percent at end-December 1994 and 71 percent at end-May 1995. Broad money growth did not increase and remained virtually unchanged as net foreign assets expansion slowed.

These policy changes appear to have been a trial balloon for the BOS to see if lower domestic interest rates and a less stringent monetary stance would solve their capital inflow problems. The outcome was surprisingly positive and reinforced the BOS’s policy stance. As expected, the pressure for nominal appreciation of the exchange rate did subside and the exchange rate stabilized, notwithstanding the slowdown in BOS sterilized interventions. A decline in both nominal and real interest rates occurred and capital inflows begin to ease. The BOS also imposed inward capital inflow controls in February 1995. These controls do not appear to have contributed to a decline in capital inflows during February-Nay, 1995, although it is impossible to quantify the impact. 1/ After several months core inflation declined substantially to below ½ percent per month during April-July 1995.

5. Instruments of monetary Policy

The BOS was extremely active on a broad number of fronts modifying and refining its refinancing facilities on a regular basis and selling its own bills to achieve an array of policy goals. While its main objective was to implement its reserve money program, the secondary goals for the refinancing facilities and sale of bills included inducing depreciation of the exchange rate, providing liquidity support to rehabilitation banks, and influencing interest rates. For the most part, the BOS did not use interest rates to ration access to its refinancing facilities or determine the volume of its sales of bills. Instead, the BOS established conditions for access to its refinancing facilities requiring BOS securities as collateral and requiring banks using its facilities to purchase foreign exchange from exporters at a specified exchange rate. The BOS auctioned either the quantity of foreign exchange to be purchased from exporters or the exchange rate for the transactions rather than auctioning access to its refinancing facilities via interest rates.

Reserve requirements were unchanged during 1993 and 1994 and restructured in April 1995. Reserve requirements were decreased on demand deposits, increased on medium-term deposits, lowered on longer term deposits, and eliminated altogether for deposits of more than one year maturity. 1/ For a period from November 1994 to June 1995 the BOS imposed excess reserve requirements on banks paying higher-than-average interest on tolar deposits (see section 8 below). There are no reserve requirements on foreign exchange deposits, instead there are foreign exchange cover regulations, these were raised slightly in 1995. 2/

BOS bills were an integral part of monetary policy. At the time of monetary independence the BOS possessed no Government securities in its portfolio to implement monetary policy as only limited Government securities existed. Tolar currency was originally issued in exchange for dinars of the FYR in October 1991. With BOS policy oriented towards sterilizing a portion of its foreign exchange purchases beginning in mid-1992, the BOS undertook to sell its own bills. Three basic types of BOS bills have been sold: tolar denominated, foreign exchange bills, and twin bills (with tolar and foreign exchange components). The foreign exchange bills are purchased with foreign exchange, have maturities from 60. to 120 days, and since May 1994 from 180 to 360 days. Tolar and twin bills are purchased with tolars. Tolar bills have a maturity of 2 to 60 days; twin bills have maturities of 3 to 6 months. Half the twin bills are repaid in foreign exchange. Bills are made available on-tap rather than sold by auction. Interest payments were indexed to inflation or exchange rates (see below). Foreign exchange bills were sold mostly to banks to help them meet foreign exchange cover regulations.

In June 1994 the BOS introduced a new bill which specified a nominal interest rate (17 percent) for the first time, with a maturity of six months. Removable warrants were attached to this bill that stipulated that if inflation were above a designated level (13 percent), the warrants could be used to purchase the next issue of foreign exchange bills or tolar bills at a discount. These warrants increased the effective rate of return and, therefore, were a substitute for explicit indexation or a higher nominal interest rate. The discount was based upon the deviation of actual inflation from the designated level when used to purchase tolar denominated bills, or alternatively the deviation of the actual DM exchange rate depreciation from designated inflation. Thus, the warrants provided protection against higher-than-designated inflation or exchange rate depreciation that was less than designated. When the original bills matured in December 1994, another series was issued at only one third the original volume. The third issue of warrants was offered in June 1995 carrying a 9 percent per annum rate of interest and inflation designation of 0.6 per month (7.4 percent per annum).

The major refinancing Instruments consisted of Lombard loans for which BOS bills were used as collateral; liquidity loans which served as a window of last resort for rehabilitation banks and banks that are net borrowers in the interbank market; short-term loans against foreign exchange bills; and foreign exchange operations. 1/ Foreign exchange operations consisted of outright purchases, repurchase agreements (REPOs), and temporary purchase of BOS foreign exchange bills. 2/

The Lombard interest rate was set well below interbank market rates and instead used to influence the exchange rate (Table 52). They were nominal interest rates while all other BOS interest rates were indexed to inflation in the case of tolar denominated facilities and bills or exchange rate depreciation in the case of foreign exchange denominated bills. For liquidity loans, there was virtually no spread between refinancing rates and interest payments on BOS bills. Instead of using interest rates to ration refinancing facilities, access to the facilities was regulated by conditions designed to influence the exchange rate. The BOS stipulated that banks must purchase a specified amount of foreign exchange from exporters within a limited period and at a specified exchange rate in order to use the Lombard facility, special 14-day liquidity loans to rehabilitation banks, normal REPOs, and temporary foreign exchange purchases. 3/ The BOS auctioned these instruments based on either the exchange rate or the amount of foreign exchange to be purchased. In this manner the BOS designated its immediate objective for. the exchange rate and assured that tolars were supplied at this rate.

There were several facilities and policies that supplied liquidity to the rehabilitation banks. These banks were partially relieved of fulfilling reserve requirements during April 1993 till mid-1995. 1/ The BOS also set up special facilities for these banks, including one-day and 14-day liquidity loans for which interest rates were often below time deposit rates.

To supplement sterilization policy, the BOS imposed capital controls in February 1995. These required enterprises that borrowed abroad (less than 5 year loans) to either immediately use the proceeds to finance imports or deposit 40 percent of the tolar equivalent of proceeds into unrenumerated tolar accounts. Longer term loans were not subject to any restrictions. The main impact of this regulation appears to have been to induce enterprises to arrange longer maturity loans.

6. Inflation and demand for money

The reasons why core monthly inflation seemed to stagnate during 1993–94, after having fallen so sharply in 1992, is a difficult question to answer. Tracing the movements in the nominal exchange rate, monetary aggregates, and wages-administered prices provide a reasonably plausible explanation for the observed inflation patterns in 1992 and 1993. The lack of progress in reducing inflation in 1994 is, however, not well explained by these variables. The sudden drop in inflation during 1995 appears to be attributable to the stabilization of the nominal exchange rate.

During 1992, a number of economic variables combined to induce a large decline in monthly inflation. The nominal exchange rate depreciated slowly—RPI inflation during 1992 was 93 percent while the nominal exchange rate depreciated 38 percent—and administered prices increases were below core inflation. The accumulated impact of these trends led to a sharp drop in inflation up to mid-1992 and another drop at the end of 1992. It is hard to characterize the impact of monetary aggregates. Although real money aggregates increased sharply, this increase appears to have been an accommodation of an increase in demand for money.

During the second half of 1992, several policy shifts occurred. Although these shifts did not offset the inflation gains made in 1992, they had a negative impact on inflation reduction in 1993. After a short period (February-June 1992) of nominal exchange rate stability against the DM, the nominal exchange rate began to depreciate again during the second half of 1992. In 1993, the exchange rate depreciated by 27 percent while inflation was 23 percent during the year. Adjustments in administered prices contributed 10 percentage points to inflation in 1993. 2/ Growth in real narrow money declined during 1993, growth in real M2 accelerated sharply, and real broad money grew at about the same pace in 1993 as in 1992. Again, these real increases in monetary aggregates appear to have reflected growth in real money demand.

In 1994, the nominal exchange rate depreciated more slowly—only a 4 percent depreciation took place in 1994 compared to inflation of 18 percent. Growth in real money aggregates continued in 1994 but at about half the pace of 1993. Administered prices increased at half the pace of 1993 contributing to a 5 percentage point decline in the overall inflation rate.

A substantial decline in inflation occurred beginning March-June 1995. Core inflation fell to an average of 0.4 percent during March-June 1995 compared to 1.2 percent per month during the same period in 1994. The main factor appears to be the exchange rate. The nominal exchange rate against the DM not only stabilized but appreciated by 5 percent from December 1994 to May 1995.

A possible explanation for this pattern of inflation reduction is that the exchange rate depreciation coupled with expectations and a highly indexed economy hindered progress. Evidence appears to point to the exchange rate as the main determinant of inflation. Granger causality tests confirm that the exchange rate influences inflation (Appendix I). The time lag is short, two months. As expected, the impact of the exchange rate on core inflation is much stronger than overall inflation. The reverse causality of core inflation on the exchange rate is found to oscillate, the effect is positive and significant for one month lag and negative and significant for two months lag. This bidirectional causality is to be expected given the formulation of the reserve money program and the real exchange rate rule that prevailed during most of the time period under consideration. Additionally, Granger causality tests indicate bidirectional causality between reserve money and the exchange rate. Granger causality tests indicate only a weak causal relationship exists between narrow money and inflation in both directions. 1/ No causality was detected between inflation and tolar time deposits, M2, foreign exchange deposits, or broad money.

Econometric estimates indicate that the demand for narrow money is stable and adjusts rapidly to changes in price levels. Real narrow money also increases with real household income, is inversely related to real interest rates, and increases with the exchange rate. 1/ Another interesting finding is that the time trend in this difference equation is positive and significant. This points to a sustained increase in demand for money and remonetization of the economy that is not explained by the variables included in the econometric equation. This interpretation is supported by the rapid rise in tolar time deposits relative to foreign exchange denominated deposits. In 1994, foreign exchange denominated deposits increased only 12 percent in DM terms (20 percent in tolar terms), while tolar time deposits rose by 80 percent. Econometric tests of M2 indicate an unstable equation, which was expected given the high degree of indexation of tolar time deposits. The econometric estimates of broad money indicate a stable function. However, the indexation of tolar time deposits to inflation and of foreign exchange deposits to the exchange rate renders broad money supply highly endogenous and thus an unsuitable nominal anchor.

7. Capital inflows and sterilization

Balance of payment surpluses presented a major complication for management of monetary policy, leading to a large accumulation of international reserves as banking system holdings of foreign assets quadrupled to US$3.3 billion during mid-1992 to mid-1995. The nature of the inflows varied over the period (see Chapter on External Sector Developments). During mid-1992 to mid-1994 the BOS resisted pressures for an exchange rate appreciation through active sterilization. The BOS directly increased its net foreign assets and induced banks to increase their net foreign assets, successfully implementing an unannounced real exchange rate rule. During 1992 this required extensive sterilization. The BOS increased its international reserves mainly by selling BOS bills, leading to a large increase in BOS net foreign assets and a decline in BOS net domestic assets (Chart 11). In 1993 when external surpluses moderated, BOS sales of bills declined and BOS net foreign assets rose at a more moderate rate. In 1994, net foreign assets of the BOS increased rapidly and sales of BOS bills rose leading to a fall in net domestic assets of the BOS. The net foreign assets of the banking system increased rapidly during 1992, fell during 1993, and increased again during 1994 (Chart 12). Net domestic assets of the banking system were almost constant in 1994 as broad money growth was entirely due to net foreign assets increases.

CHART 11
CHART 11

SLOVENIA: DEVELOPMENTS IN RESERVE MONEY

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by the Slovenian authorities
CHART 12
CHART 12

SLOVENIA: DEVELOPMENTS IN MONETARY AGGREGATES

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by the Slovenian authorities.With increases in the stock of outstanding BOS bills, increased interest expenditures virtually eliminated BOS profits in 1994. According to the BOS Annual Report, BOS profits were 0.8 percent of GDP in 1992, they tripled to 2.5 percent of GDP in 1993, and nearly disappeared in 1994 falling to 0.1 percent of GDP. Two-thirds of the 1993 profits of the BOS were unrealized capital gains resulting from the large currency depreciation. The profit decline in 1994 reflected the high sterilization costs and the modest level of nominal exchange rate depreciation (7 percent).

BOS balance sheet changes indicate the extent to which the BOS was oriented towards sterilization (Table 50). Net foreign assets increased by SIT 110 billion from end-1993 to May 1995 while reserve money increased by only SIT 35 billion, leading to a considerably drop in net domestic assets. The decline in net domestic assets was mostly brought about by net sales of BOS bills of SIT 60 billion from end-1993 to May 1995, of which 60 percent were foreign exchange bonds.

8. Interest rates

Interest rates—both deposit and lending rates—have been positive in real terms throughout the transition period 1991–94. The main reason was the indexation of financial assets which focused attention on the real component. The rate of return on tolar instruments was indexed to the previous month’s domestic inflation, plus a real interest rate premium (r); thus the ex-post rate of return was actual inflation in the previous month plus (r). These deposits could also indexed to exchange rate movements with a real rate of return (d). The real rate of return en foreign currency indexed assets was about 25 percent below the real interest rate on tolar indexed assets during 1993–94; in 1995 the gap between (r) and (d) closed (Chart 13). Moreover, the ex-post rate of return on tolar assets has exceeded the ex-post rate of return on foreign exchange deposits from mid-1993 to mid-1995. Inflation has exceeded depreciation of the tolar-DM exchange rate throughout this period (except for November 1993), providing a higher nominal return to assets indexed to domestic inflation than to assets indexed to the DM.

CHART 13
CHART 13

SLOVENIA: INTEREST RATE DEVELOPMENTS

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by the Slovenian authorities.

The rate of return on tolar assets relative to foreign assets has three elements—country or credit risk, exchange rate risk, and interest rate differentials. Interest rates on DM deposits in Slovenia were below German deposit rates by as much as 2 percentage points during 1993 and most of 1994; this gap narrowed as German rates fell. This data would suggest that banks in Slovenia did not have to pay a premium to attract DM deposits, which is a possible measure of country/bank risk. Thus, country credit risk does not appear to be a major contributing factor for interest rate differentials.

Exchange rate risk may also be a factor explaining interest rate differentials. During the period mid-1992 to mid-1994, the nominal exchange rate depreciated against the DM roughly in line with inflation differentials. Interest rate differentials, therefore, incorporate expected movements in the exchange rate. As the nominal exchange rate gradually stabilized from mid-1994 to mid-1995, interest rate differentials narrowed.

An alternative reason for the high rates of return on tolar assets may have been the tight monetary policy. Interest rates are generally regarded as a barometer of monetary policy. Their high real level could imply that monetary policy was restrictive in Slovenia during the period 1992–94. Certainly, domestic interest rates dropped during mid-1994 to mid-1995 as domestic credit conditions were eased.

Another explanation for high real domestic interest rates can be found in the structure of the banking system and overall financial markets. High tolar deposit rates may reflect bank competition for tolar deposits to maintain market shares at the expense of maximizing profits. Some banks relend their tolar deposits on the interbank market to banks short of liquidity, such as the rehabilitation banks, at an interest rate only slightly above deposits rates. 1/ Other banks purchased foreign exchange assets for which foreign interest rates were below the domestic interest rate, even as the exchange rate crawl slowed.

The BOS was concerned that interest rates were too high and took several actions to lower then. Lombard interest rates were kept below market rates throughout and were lowered when inflation fell. 2/ During 1993–94, real interest rates on liquidity loans were gradually lowered from 10 percent to 6 percent. The BOS considered the main reason for the high tolar deposit rates was competition for tolar deposits by banks. Effective November 1994 (enacted June 1994), additional reserve requirements were imposed on banks whose interest rates on short-term tolar deposits exceeded the average by the banking system. This regulation was rescinded effective end-June 1995. In December 1994, the BOS brokered a voluntary agreement among the banks to reduce their short-term tolar deposit rates. A second agreement came into force in April 1995. 3/ Beginning in the fourth quarter of 1994, real interest rates on the interbank market began to recede quickly. Commercial bank real interest rates did not decline until the second quarter of 1995. Combined with the decline in inflation, nominal interest rates were cut to one third of their previous level. 4/

The reported spread between lending and deposit rates of around 12 percent during 1992–94 were also quite high. 1/ The deposit and lending rates reported by the BOS represent the mid-point between the highest and lowest interest rates offered by banks rather than an appropriately weighted average. Thus, the implied interest rate spread may have an upward bias as most transactions might be with preferred customers who receive both higher deposit rates and pay lower lending rates. In addition, lending rates exhibit very little variation with maturity, while deposit rates have a substantially greater range. For example, in 1994, the real lending rates had an average range of 1 percent compared to 15 percent for real deposit rates. Thus, depending on the liquidity of borrowers and the maturity composition of deposits the actual spread could be substantially smaller.

Support for the view the spreads are narrower than reported is provided by the low level of net interest income reported by banks in 1994 (2 ½ percent of assets). Net interest income is not a pure measure of domestic interest rate spreads as it mixes net interest income on both tolar and foreign exchange interest rates. Available information does not permit a separation of these two sources of net interest income. Banks also had a large share of non-interest bearing assets (see Chapter on Structural Reforms), principally because they have a high desired capital asset ratio. Non-performing loans are no longer a significant reason for spreads either.

IV. Fiscal Policy

Slovenia has achieved impressive results in building a fiscal apparatus and maintaining fiscal balance since 1991. Independence created an administrative vacuum because of the emphasis placed on federal and local administrations in the former Yugoslavia. The new fiscal administration was able to exert control on fiscal expenditures swiftly.

l. Fiscal operations since independence

Slovenia was faced with many of the problems encountered by other transitional economies. Transitional economies from Eastern and Central Europe experienced substantial fiscal deficits which, as a percent of GDP, reached double-digit levels in Albania, Bulgaria and the Slovak Republic. These deficits stemmed primarily from lower revenues as a percentage of GDP. 1/ In contrast, Slovenia was able to maintain revenue collection and recorded a string of fiscal surpluses from 1991 to 1993 (Table 53). Generous welfare spending and a costly restructuring of the banking and enterprise sectors boosted expenditures, while disruptions in tax administration threatened revenues. However, Slovenia began the transition with a relatively strong fiscal position since substantial transfers to the federal budget of the former Yugoslavia, which amounted to on average about 9 percent of GDP during 1988–1990, were eliminated. These fiscal savings were used to create a fiscal administration and finance structural reforms.

a. Taxation

All transitional economies experienced a decline in revenues relative to GDP, bordering on collapse in certain cases. Hungary and Poland experienced revenue declines equivalent to 6–8 percent of GDP; in Bulgaria and Romania revenue losses reached 20–30 percent of GDP. The causes of these falls were twofold: an erosion of the corporate tax base and a decline in effective tax rates, reflecting tax evasion, weak tax administration, or policy decisions.

In Slovenia, by contrast, budget revenues which averaged 44 percent of GDP during 1988–90, exceeded 46 percent of GDP on average during 1991–94. Two factors were instrumental in this stronger revenue performance relative to other transitional economies. First, Slovenia has relied more heavily on wage and consumption taxes, rather than on corporate taxes or dividends from state-owned firms. Second, this favorable tax structure has been effectively implemented by a well-performing tax administration, the Agency for Payments, Supervision and Information (APPNI).

(1) A high reliance on indirect and wage taxes

In Slovenia, the tax burden is high and borne mainly by wage-earners and consumers. As a corollary, revenues from other taxes (e.g., corporate or property taxes) are lower than in other countries (Table 58). The personal income tax (PIT) is de facto a wage tax. In theory, this tax is fairly broad-based. 1/ Wages and salaries represented 73 percent of registered household incomes (excluding social transfers and capital income) in 1994 and 87 percent of the tax base in that year. By contrast, the taxation of profits from individual entrepreneurs or small businesses amounted to only 2 ½ percent of PIT revenues in 1994.

Relatively low taxes on non-wage incomes reflects a slow adjustment of the tax system to the new economic environment and has also created a distributional concern—wage earners and other occupational groups are not on an equal footing for tax purposes. This is due, in particular, to lenient treatment of fringe benefits and of incomes of the self-employed. Fringe benefits are appealing because they are often undervalued and they are also legally exempt from social security contributions. 2/ For the self-employed, the assessment of certain categories of income for tax purposes is somewhat blurred (e.g., professionals, consultancy work).

Introduction of taxes on capital gains and interest income has been postponed until 1997, except for capital gains on real estate transactions and dividend income. These exemptions create other loopholes on the income tax base. In addition, as pointed out in IMF (1993), such tax exemptions also affect the return on financial and non-financial assets, encouraging financial decisions based on tax arbitrage. 3/

The undue reliance on wage taxation has obvious drawbacks. In order to generate high revenues, marginal tax rates are fairly high and increase steeply to SO percent. As Slovenia’s income distribution is skewed towards low income (Table 9), households bearing the top marginal rates are few in number. In addition to five tax brackets, the PIT provisions are further complicated by numerous exemptions.

Table 9.

Proportion of Tax-Payers for Each Marginal Tax Rate

(In percent)

article image
Source: Slovenian authorities. Note: AW = average wage in the economy.

Social Security contributions are earmarked for Pension and Health Funds and for unemployment benefits (Table 10). According to the law, contributions provide the financing of legally defined entitlements for pensioners and health insurees. Contribution rates have declined since 1992, and were set at 44.7 percent of the net wage in 1994, including 31 percent for pensions and 12.7 for health insurance. Accordingly, combined with the PIT, the marginal tax rate reaches 55 percent for average-wage earners. The self-employed have more latitude concerning social security contributions since they are allowed to choose their tax rates (15 ½ or 13 ⅓ percent) and define their tax base. The self-employed can, for instance, legally define their taxable income at the minimum wage, whatever their actual income. Choice of the minimum wage would simply reduce their future pension entitlement which is relatively generous for low income groups. As a result, most of the self-employed choose to minimize their contributions.

Table 10.

Social Security Contributions, 1991–95

(In percent of net wages)

article image
Source: Slovenian authorities.

Taxes on goods and services represent another pillar of the tax system. They comprise three distinct categories: a general sales tax, excise taxes and customs duties. On average, during the period 1992–94, tax revenues from these three categories rose by 7 ½ percent, 4 percent and 3 ½ percent, respectively. Slovenia is amongst the few Eastern European economies that still rely upon a sales tax instead of a VAT. The general rate for goods is set at 20 percent with reduced rates for construction materials (10 percent), capital goods (5 percent), and basic staples (5 percent). A luxury rate of 32 percent is applicable to certain items (jewelry, luxury cars, etc.). As for services, a 5 percent tax is levied, with the exception of some financial services (3 percent). Excise taxes are low by European standards, particularly for wine. In 1994, new taxes were introduced on cigarettes and alcohol. Low rates on oil derivatives relative to other European countries attract car drivers from neighboring Italy.

Customs duties represent an important share of revenues because of the high degree of openness of the Slovenian economy and relatively high duties. A reshuffling of these taxes, required for European integration, will be implemented on January 1996. Since there is currently no tax on exports, the major impact of the planned reform is an across-the-board reduction in import tax rates. Tax revenues are expected to shrink by 2–3 percent of GDP.

As mentioned, Slovenia’s budget does not rely upon corporate tax revenues which raised on average less than 1 percent of GDP during the period 1991–94. Moreover, unlike other communist countries, the state budget did not receive dividends from state- or socially-owned firms. Therefore, total budget revenues were hardly affected when firms became unprofitable following the restructuring of the economy. 1/ In September 1994, corporate tax rates were lowered from 30 to 25 percent with retroactive impact from January 1994. At the same time, some of the numerous tax exemptions and tax holidays granted to specific sectors (e.g., agricultural enterprises, small businesses) were eliminated. Lower corporate tax rates were intended to place Slovenia on a more equal footing with Croatia whose corporate tax rate is 25 percent.

Corporate tax revenues as a percent of GDP are expected to rise at a slow pace. Indeed, the impact of improving profitability will be delayed by the carry-over of past losses. This situation may improve significantly once firms are privatized. Indeed, it is sometimes suggested that generating losses is a device used by managers to lower firms’ value before privatization.

(2) The tax administration(s)

Over the 1991–1994 period, the Slovenian tax administration has performed extremely well despite the split responsibilities between the APPNI and the Ministry of Finance (the Bureau of Public Income). The former deals with legal entities, while the latter administers taxes borne by natural persons (individuals and unincorporated businesses).

The APPNI is an independent body reporting to Parliament. It has wide ranging powers over firms, which go well beyond its tax collection responsibilities. These include, in particular, operation of the payment system—to be taken over by the BOS—and auditing of socially-owned firms. The APPNI manages and, to a certain extent, controls the accounts of all legal entities. Acting as a bank, it transfers funds between accounts in a highly effective manner—transfers are available within a day. At the same time, and unlike a bank, the APPNI has the legal power to stop certain transfers. Indeed, it performs audits of the financial statements of entities and has to ensure that certain obligations (including corporate tax payments, and pension contributions) are given priority treatment. It also deducts from enterprise accounts tax payments related to wages. To illustrate this encompassing role, the APPNI has successfully deterred fraudulent transfers of assets during the privatization process and enforced the wage agreements of 1994 and 1995.

The Bureau of Public Income has more modest objectives and powers. It was created only in 1992, when a new law brought community tax offices under a central administrative umbrella. The Bureau has suffered from an inability to impose stiff penalties on tax evaders since these can only be determined by an economic court. Consequently, tax compliance for non-wage earners has been a problem. Estimates by Slovenian authorities indicate that delayed payments represent almost 50 percent of the amount due. It might well be just a delay due, for instance, to the progressive computerization of the tax system, but it could also be indicative of a certain reluctance to pay taxes amongst non-wage earners.

There is a widespread consensus that the tax administration needs to be unified under the auspices of the Ministry of Finance. Such a reform has been envisaged since 1992, in particular, to facilitate the introduction of a VAT (see below). This unification has not yet been completed.

b. Expenditures

General government spending increased from 43 ½ percent of GDP, on average, during 1988–1990 to around 46 percent of GDP on average during 1992–94. This increase reflects costs associated with maintaining a well-developed safety net, restructuring banks and enterprises, and building of a new tax administration.

(l) Social safety net

As in many transitional economies, the budgetary costs associated with the social safety net (pensions, health expenditures and unemployment benefits) have soared since the beginning of the transition. From 1991 to 1994, social safety net costs have increased by 5 percent of GDP to 22.4 percent of GDP in 1994 (Table 11). Pensions and health insurance expenditures have been the main factors accounting for the increased cost of the social safety net. Unemployment benefits despite high unemployment peaked at only 1 ½ percent of GDP in 1993, and with lower unemployment in 1994, fell by 0.2 percent of GDP.

Table 11.

Social Transfers

(As a percentage of GDP)

article image
Source: SORS, various publications.

Including old-age, disability and survivors.

FY1995 Budget.

The causes for the rising cost of Slovenian pensions are analyzed in Appendix II. Early retirement, as a substitute for unemployment benefits for older workers, played a key role in increasing pension costs. Early retirements led to a substantially greater number of pension recipients and to a decline in activity rates of older workers. This eroded the number of contributors to the pension system. Slovenia has also shown a strong commitment to maintaining the standard of living of pensioners relative to workers; pension benefits have been tightly linked to the average wage.

The cornerstone of Slovenia’s health policy has been to maintain full coverage for the entire population. By law, Slovenian citizens are entitled to a basic package of health insurance benefits, either directly or as family members. This principle also applies to the unemployed. However, benefits and coverage vary across age groups and according to the specific disability/treatment. To contain expenditures, Slovenia decided in 1992 to evolve from a state-provided health care system to a health insurance system. The agency in charge of public health, the Health Institute (HI), is responsible for the financing, but does not necessarily provide directly health services. The HI has initiated an across-the-board reduction in expenditure growth by encouraging competition and relying on private health service providers selected through a system of tenders. It is also targeting fast-rising expenditures (e.g., drug consumption, orthopaedic expenditures) for remedial action.

In 1994, 90 ½ percent of the health services were covered by the basic insurance compared to 98 ½ percent in 1992. The uncovered fraction can be supplemented by voluntary insurance, public or private. The HI itself offers various packages and has enrolled 1.2 million members in the voluntary program in 1994. In addition, private companies also may insure co-payments. Currently, private companies have sold less than 50,000 insurance policies.

The compulsory component of health insurance is financed mainly by contributions levied on employers and employees (6.1 percent of net wages each); employers contribute an additional 0.5 percent for occupational illnesses and work injuries. For other groups, insurance is paid by the Pension Fund municipalities, underlining the solidarity element of the system.

(2) Restructuring cost

The budgetary costs of restructuring programs rose until 1993 and then declined slightly in 1994 (Table 12). The most remarkable feature of these costs in Slovenia has been the emphasis placed upon active labor market policy (e.g., retraining policy). The government has also intervened to facilitate enterprise assistance and bank restructuring. Various sectors benefitted from government assistance, including steel mills, the mining industry, and more recently the railways.

Table 12.

Government Expenditures on Restructuring Programs of Enterprises and Banks, and for Social Safety Net 1992–95

(As a percent of GDP)

article image
Source: Slovenian authorities.

Other than unemployment benefits.

In 1993, the MOF recapitalized three major banks, which represented more than 60 percent of the banking sector assets. Non-performing loans on balance sheets of these banks were replaced by 30-year government bonds (indexed to the deutsche mark) and earning 8 percent interest (Table 13). Restructuring bonds represent the bulk of the government domestic debt. To spread restructuring costs over a long period of time, the government has issued long-duration bonds. Therefore, the budgetary cost barely exceeds interest payments.

Table 13.

Long-Term Government Securities

(Outstanding - 12/31/94)

article image
Source: Slovenian authorities.

In millions of deutsche marks.

(3) Fiscal administration

Prior to independence, the republican fiscal administration was meager and was squeezed between self-managed communities and the federal administration. Self managed communities accounted for approximately two-thirds of public expenditures and were quite independent. With independence, the federal administration of the former SFRY vanished and its activities (e.g., defence) were taken over by the new Slovenian administration. The system of self-management has largely disappeared and the local governments play only a minor role. Resources have come increasingly under the central government’s control. Between 1992 and 1994, the local governments accounted for 5 ¼ percent of GDP or approximately one-ninth of total general government expenditure. As a result, their share of tax revenues has diminished.

In order to meet the new obligations created by independence, the number of government employees rose by 12 percent between 1991 and 1994. The bulk of these new employees were utilized by the central government administration. In addition, a few thousand employees from non-profit organizations (e.g., university teachers, research, health care) or local governments were transferred to the central government’s payroll.

To attract and retain talented civil servants, Slovenia has granted high salaries, although their relative position in the wage spectrum of the economy has deteriorated since the beginning of the transition (Table 14). Nonetheless, in 1994, wages in the public sector surpassed enterprise wages by approximately 30 percent. A large fraction of this differential reflects higher living costs and higher qualifications. 1/ To some extent, the need to build up a new administration quickly has been an underlying rationale for maintaining this differential. However, over time the authorities will need to be satisfied that the optimal allocation of skilled labor is achieved between the public and private sectors.

Table 14.

Wages and Employment in the Public Sector

article image
Source: Slovenian authorities.

Growth rates compare 1995 Q1 to the 1994 average.

Ratio of public sector wages to private sector ones.

c. Fiscal stance

During 1988 and 1990, Slovenia’s budget, including transfers to the federal budget of almost 9 percent of GDP, was balanced. After independence, these transfers were used to cover additional spending for the safety net (4 ½ percent of GDP) and economic restructuring (2 ½ percent of GDP). In addition, the new administration has also increased payroll expenditure by 1 percent of GDP and other spending (mainly for defence and new infrastructure) by 3 percent of GDP. Higher taxes (2 ½ percent of GDP) more than compensated potential imbalances, resulting in a small surplus—averaging ½ percent of GDP—over the period 1991–94. Overall, the fiscal stance has been neutral during 1991–94.

With regard to the components of the Budget, however, there has been a progressive deterioration in the financial balance of the Pension and Health Funds. This was largely counterbalanced by an improvement in the overall surplus of the central government. As a result, for FY 1994, the overall budget deficit of the consolidated general government amounted to 0.2 percent of GDP (against 0.8 percent in the Budget Law), including surpluses for the central government (0.4 percent of GDP) and local governments (0.1 percent of GDP) and deficits in Pension and Health Funds (0.7 percent of GDP).

In 1994, the stance of fiscal policy changed as the budget moved from a surplus into deficit—a swing of about 1 percent of GDP even as real growth accelerated to 5 ½ percent. Despite actual growth being higher than forecast, the scope for automatic stabilizers was reduced by the government’s actions both on the revenue and expenditure sides. Higher-than-expected real growth boosted tax revenues, which were 2 ½ percent of GDP higher than expected although unchanged from the 1993 outcome. Strong collection of sales taxes and customs duties were chiefly the reason for this over-performance. The government also budgeted reduced social security contributions in 1994. Thus, health insurance and unemployment contributions declined by 1 ½ percent of GDP relatively to an unchanged contribution scenario. This is consistent with the government’s medium-term strategy to reduce wage taxation.

In 1994, expenditures edged up by ½ percent of GDP to 47.3 percent. Pension and disability expenditures increased by 0.7 percent of GDP due to a continuing influx of early retirees and indexation of benefits upon real wages. A pressing need for new infrastructure investments also led to substantial capital expenditures (up by 0.8 percent to 3.1 percent of GDP). Simultaneously, in order to contain the budget deficit, subsidies to the private sector were pared back by ½ percent of GDP. By year-end, however, central and local governments took advantage of the higher-than-expected tax revenues to increase expenditures substantially.

Although budget figures indicate a sharp reduction in public sector wages in 1994, this is mainly due to a delay in wage payments. Starting in May 1994, the payment of wages was pushed back by about a week, to the 6th or 7th of the following month. Thus, December 1994 wages which were only paid in early January 1995 and were not included in the fiscal out-turns for 1994. This accounting convention created a one-time reduction in wage expenditures in 1994, amounting approximately to 0.6 percent of GDP. If one eliminates this effect, payroll costs would have been stable around 5.2 percent of GDP. In addition, during the last quarter of 1994 civil servants were granted a 22 ½ percent rise in average wages which would burden the 1995 Budget.

Fiscal policy, in addition to its procyclical stance in 1994, also had a pronounced seasonal pattern owing to a lengthy budgetary process. Since independence, parliamentary passage of the budget has been increasingly delayed. The budget vote took place in April, 1991, and slipped to May, 1994; the 1995 budget was passed in late June. As the Cabinet is granted by law some latitude over the timing of budgetary debates and discussions can be postponed until a political consensus is achieved. Thereafter, under existing parliamentary procedures, budget debates require at least 85 days before the final vote takes place.

The timing of Parliament’s approval of the budget largely determines the seasonal patterns of budgetary expenditures. To alleviate potential disruptions, the previous year’s total expenditures are authorized, prorated until the new Budget Law is passed. During the high inflation period, authorized expenditures were determined in real terms. This is no longer the case, which imposes a more stringent constraint upon non-mandated expenditures such as investment. Expenditures on pensions and health insurance are defined separately and are unaffected.

The timing of budgetary process induces a marked seasonal pattern in government expenditures (Chart 14). Typically, an overall surplus is recorded during the first semester, and expenditures rise sharply during the second half of the year. Moreover, during the second half, ministries also have an incentive to exhaust their allocated budgets, making advance payments in preparation for a period of stringency, during the first semester of the following year. The law facilitates this behavior by allowing for the inclusion in December expenditures of some expenditures which only occur in the following year. Hence, a marked deficit in December each year.

CHART 14
CHART 14

SLOVENIA: SEASONAL PATTERNS IN GOVERNMENT EXPENDITURES

(in billions of tolars)

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by the Slovenian authorities.

2. Fiscal reform

As the 1995 Budget indicates, past trends are likely to continue. Analysis of this budget reveals that major dilemmas confront fiscal policy. The government needs to reduce the level of taxes and modify their composition to alleviate the tax burden shouldered by wage earners and to facilitate EU integration.

a. 1995

For 1995, the projected deficit (0.5 percent of GDP) for the general government also corresponds to a surplus for the central government (0.4 percent of GDP) and for the local governments (0.1 percent of GDP) and a deficit for Pension and Health Funds (0.9 percent of GDP). In order to combine lower taxes with a roughly balanced budget, considerable expenditure discipline will be required.

The 1995 Budget Law envisages lover revenues than in 1994 by ½ percent of GDP, to 43.2 percent of GDP. In fact, developments during the first five months of the year suggest that annual revenues may exceed projected levels. Indeed, revenues were running ahead of expectations in almost every tax category during the January-May period. Growth of personal income tax has surpassed the initial budget forecast by about 6 percent, due to two factors. One, budget forecasts assume a relatively low elasticity of taxes to households’ income (the implicit elasticity is close to 1) despite the progressivity of the PIT. Two, wage inflation has been somewhat higher than expected through the first five months of 1995. This also led to higher social security contributions. Corporate taxes are buoyant despite the retroactive tax cut from 30 percent to 25 percent. Firms had paid the 1994 tax installments based upon the former 30 percent rate, thus the final payment in March 1995 was meager. Afterwards, in April and May, revenues grew robustly, indicating marked improvements in firms’ profitability. Given current trends, tax revenues are likely to exceed the budget forecast.

The 1995 budget attempts to rein in government expenditures, as part of a longer term effort to contain expenditures below 44 percent of GDP by 1997. The 1995 budget is a first step in this direction. Expenditures are planned to decline to 46.2 percent of GDP in 1995, compared to 47.3 percent in 1994. The stabilization of the restructuring costs is expected to facilitate this expenditure reduction.

In 1995, central government wage expenditures are expected to grow by more than 60 percent. Reasons for this increase are manifold. First, as already explained, one month of 1994 salaries were shifted to 1995, thereby increasing wages by about 9 percent in 1995. Second, the transfer of civil servants from non-profit organizations or local governments’ budget to the State budget may add another 15 percent to payroll costs. Third, wages are still growing rapidly. Indeed, the protracted effect of the 1994 wage agreement in the public sector has increased civil servants’ salaries by about 15 per cent during the first quarter of 1995, far above the inflation rate.

Pension expenditures were envisaged in the budget to decline slightly to 13.6 percent in 1995. However, this projection has been jeopardized by Parliament’s rejection of proposals to diminish benefits. To alleviate stringent liquidity problems, the central government has increased transfers to the Pension Fund, earmarked for the payment of war veterans and farmers pensions. Liquidity would also be provided more directly through a repurchase agreement of securities held by the Pension Fund.

Increasing central government transfers to the Pension Fund call into question a fundamental tenet of the Pay-As-You-Go (PAYG) Pension Fund: its budget is supposed to be financed by contributions not through budgetary transfers.

Overall, after five months, the general government has accumulated a surplus representing SIT 8.5 billion against SIT 16.8 billion for the same period in 1994. Yet, this does not indicate a deterioration of the general government’s fiscal position; it reflects the delayed payment of wages in May 1994 and reduced corporate tax revenues in March 1995.

b. The need for tax reform

In 1992, the Slovenian government mapped out an ambitious tax reform program. One of the major goals outlined in this plan was to facilitate Slovenia’s integration within the EU. Although this objective is still high on the agenda, other preoccupations have emerged. In particular, the reduction of wage taxes has become an issue of growing importance.

(1) Consumption and wage taxation: recent reforms and main issues

The authorities have become increasingly concerned that excessive wage taxes may threaten Slovenia’s competitiveness particularly of low value-added activities. As a result, in addition to support for specific producers, 1/ the government has embarked upon a long-term reduction in wage taxes. Obviously, the success of its strategy depends upon its ability and its willingness to control pension and health expenditures (see below). On the tax side, the government’s approach is multifaceted. It includes some minor changes in the structure of the PIT. In the medium term, the introduction of a VAT is considered essential, since it would permit to increase consumption taxes and lower taxes on wages. Such a strategy, however, is faced with considerable technical difficulties, and raises various distributional issues.

In the short run, minor changes in the tax base and tax rates for the PIT and social security contributions have taken place. In 1995, tax brackets were modified to achieve lower rates for low-income earners and higher ones for the high-wage earners. The new rates are as follows.

Table 15.

Modified Personal Income Tax Rates 1995

article image
Source: Ministry of Finance. Note: AW = average monthly wage in the economy.

To expand the tax base, exemptions are limited to 11 percent of the average wage. The effective rate is, therefore, close to the marginal rate. Between 1991 and 1994, other tax reliefs were progressively lowered from 10 percent to no more than 3 percent. In order to close some loopholes, new taxes have been introduced on certain incomes (e.g., contractual work, which used to be a vehicle to avoid paying social security contributions).

Introduction of a VAT requires substantial changes in the organization and creation of a unified tax administration. Indeed, a major advantage resulting from a VAT-based system is that it provides the tax administration with a much better knowledge about financial flows between firms if VAT and corporate tax administration cooperate. Therefore, it enhances the effectiveness of tax controls on firms and reduces the scope for tax evasion. Complex tax issues and concerns about the risk to tax collections have been pivotal in the administration’s decision to postpone the introduction of a VAT until 1997.

A new VAT will face limits in the amount of incremental revenues it might yield. A small open economy like Slovenia is significantly exposed to cross-border shopping and cannot impose significantly higher consumption taxes than its neighbors. Consumption taxes are already high: the ratio of sales tax (including excise) to private consumption reached 25 percent in 1994. Admittedly, a VAT would broaden the tax base since certain categories of investment and public consumption also bear taxes. Still, most observers calculate that an effective rate around 16 to 18 percent is required for a budget neutral switch from a sales taxes to VAT. This leaves little scope to increased revenues stemming from the change to the VAT to finance lower custom duties or to introduce a lower rate for basic items.

Even if it were possible to increase consumption taxes and lower wage taxes, the impact on wage cost and competitiveness could be limited. This argument hinges directly on the functioning of the labor market and developments in real net wages. In a competitive closed economy where workers consume their wage income, a revenue-neutral switch from a (flat-rate) wage tax to a (flat-rate) consumption tax does not have any impact. Since consumption is assumed to equal wages, the tax base is unaltered and nominal wages increase to compensate for the tax-induced increase in the cost of living. Ultimately, real net wages remain unchanged and the impact on the real economy is nil.

Any significant impact from a switch from wage taxation to a VAT stems from certain distributional shifts, inadequately described under the previous assumptions. The most relevant ones are as follows.

  • - Labor market imperfections: Some inertia in wage determination may lead to a reduction in real net wages following the introduction of a VAT. However, in Slovenia wages are highly indexed, and with a short lag to prices.

  • - Wage earners vs. other income earners: Consumption in Slovenia is much higher than wages. In 1994, households’ purchases of goods and services represented more than twice the amount of wages and salaries that they received. Thus a VAT would broaden the tax base and, in particular, pensioners and the self-employed would bear higher taxes. A larger tax base would also allow a reduction in marginal tax rates, thereby limiting their distortionary effect. From this angle, the main advantage of a VAT is that it could facilitate the taxation of certain groups which evade the PIT.

  • - Replacing wage taxes by a VAT would also favor the manufacturing sector against the service industry for three reasons. First, exporters bear wage taxes not a VAT; the opposite holds for importers. Hence, introducing a VAT is akin to a devaluation. Second, a VAT also differs from the existing sales tax in that, to avoid cascading, existing tax rates on all services are fairly low (5 percent or below), to the benefit of certain segments of the service industry, with high value-added and low intermediate consumption. A VAT would eliminate this advantage. Third, investment goods would become tax exempt to the benefit of the capital-intensive industries in the manufacturing sector.

  • - The redistribution between households with different incomes is probably less relevant. A European-style VAT would eliminate the 32 percent tax rate on luxury goods, thereby limiting its redistributive effects. To alleviate this potential concern, one may note that luxury tax revenues amount to less than 4 percent of the general sales tax revenues. Thus, a consumption tax is probably the best tax instrument to achieve redistributive goals. One may consider instead specific taxes on luxury cars or taxes on wealth.

(2) Controlling expenditures by setting proper microeconomic incentives 1/

In the long run, the most significant challenge to a sound fiscal policy stems from its well-developed social transfers—the pension and health systems. The ageing of the Slovenian population places financial pressures on both the health care and pension systems. According to the HI, the health-related costs for the elderly (above 65) are three and a half times higher than for persons between 7 and 40 years of age. Current demographic forecasts indicate that the ratio of the two groups will increase from 26 percent to 38 percent within 15 years.

An avenue for reform is to increase the reliance on the private sector and to set proper microeconomic incentives. Concerning pensions, the following considerations are examined in the Appendix II. First, Slovenia could increase the share of fully-funded pensions, as its financial markets mature. Second, the existing incentives for early retirement should be phased out as unemployment recedes. Eliminating early retirement could more than offset the adverse financial consequences of the ageing population.

With regard to health care, Slovenia has already introduced voluntary insurance and is planning to increase its role, up to 18–20 percent by year 2000. Arguably, this will not be sufficient to cap expenditure growth. Certain costs will be extremely difficult to contain. In particular, doctors and other specialists are relatively poorly paid compared to their European counterparts. Closer ties with Europe and more competition from the private sector might produce competitive pressures for relative salaries of health care providers to rise, further increasing health costs. 2/

The main risk of the current co-insurance system is that widespread insurance diminishes incentives to reduce medical consumption. With the compulsory scheme, households bear part of the cost for health services and are inclined to reduce their consumption. A second layer, which allows for co-insurance, reduces or even eliminates these incentives if it does not screen its participants adequately—higher consumption should be matched by higher premiums. Currently, it may be argued that this is not the case. Indeed, the voluntary side of the health insurance also plunged into deficits in 1994. In more general terms, a tight screening of the population seems to contradict the HI’s policy which aims at increasing coverage to the entire population.

V. Structural Reform

1. Financial sector reform

a. Overview

The financial system in Slovenia consists of 33 banks, 11 savings institutions, and 74 savings cooperatives. The stock exchange offers shares for about 50 private companies, BOS bills, and certain issues of Government bonds. An interbank money market has existed since end-1992. The BOS and the Government have actively promoted the development of more efficient financial markets through rationalization, regulation, and supervision of banks. Rehabilitation of the large financially troubled banks has progressed.

b. The payments system

Perhaps one of the main constraints on the development of diversified financial markets is the expensive and anachronistic payments system inherited from the former-Yugoslavia. The APPNI 1/ operates the payments system on the same basis as was done prior to independence. It is also responsible for tax collection from enterprises, monitors enterprises financial performance, and collects information on transactions.

The APPNI operates several accounts for each bank and registered enterprise and all transactions—other than cash payments—must pass through the APPNI. Cash payments above a certain level are prohibited to prevent tax avoidance. The APPNI automatically deducts its transaction fees and tax liabilities from enterprise accounts based upon calculations of its staff. Thus, the APPNI operates a system of presumptive withholding taxation of enterprises in conjunction with the payments system. This system is expensive; the APPNI charges 0.1 percent on each transaction. Transactions revenues cover all the agency’s expenditures. Thus, transactions fees are used to cross subsidize other activities of the APPNI’s, creating a tax-wedge which is nearly prohibitive for short-term financial transactions—such as in the interbank market.

Technical assistance from FAD and STA, as well others including the U.S. Treasury, have counselled replacing the APPNI with a separate payments system owned by users (mainly the BOS, banks, and the stock exchange) and reverting tax administration to the Ministry of Finance. 2/ The Government plans such a replacement during 1996–97. This reform will lower fees considerably, increase the quality of banking services provided, and is expected to improve the liquidity and depth of financial markets through a boom in transactions. The regulatory and statistical functions now performed by the APPNI will need to be either distributed to other agencies or new sources of financial support will need to be found for the APPNI.

c. Interest rates and indexation issues

Immediately following monetary independence, the BOS decided to continue to index financial assets to inflation or exchange rate depreciation. The BOS calculated that with unanticipated falling inflation, indexation would lead to a faster decline in interest rates than under a regime of nominal interest rates (Bole (1995)). Accordingly, all interest rates of commercial banks, in the interbank money market, or on Government securities, BOS bills, and most refinancing rates, were indexed. The index was published by the BOS with tolar denominated assets indexed to the previous month’s inflation and foreign exchange denominated accounts indexed to the BOS’s official exchange rate (an average of the previous 2 month’s market exchange rate). Banks set real interest rates on tolar assets (r) and on foreign exchange denominated assets (d) when accepting time deposits, making loans, or transacting in the money market. Demand deposits were only partially indexed at 40 to 50 percent of the previous month’s inflation (until mid-1995 when they became fully nominal). While the rates on BOS bills and liquidity loans were indexed, Lombard and discount rates were not indexed and were kept low, generally between the interest rates on time deposits and demand deposits. The BOS wanted to offer an attractive interest rate so as to be in a position to ration use of these facilities via conditions designed to influence the exchange rate.

Both bank deposit and lending rates on tolar assets were high compared to rates on foreign exchange denominated assets. For instance during 1994 the annualized real interest rate on time deposits was 7 ½ to 11 percent (depending upon maturity) compared to 5 percent on foreign exchange time deposits. With the slowing crawl against the DM in 1994, the rate of return on tolar assets (28 percent to 32 percent) exceeded by 16 percentage points the tolar rate of return on Slovenian DM deposits (12 percent). This was higher than the uncovered interest rate differential in 1993. Consequently, foreign exchange deposits in DM terms expanded less during 1994 (12 percent) than in 1993 (33 percent). Tolar time deposits increased four fold faster (80 percent) than the rate of increase of foreign exchange deposits in tolar terms (20 percent) in 1994.

Beginning in the second quarter of 1995 the BOS adopted the goal of eliminating indexation. They began by eliminating indexation for assets below 30 days maturity and by requiring that the rates of interest on short-term deposits and loans be published in nominal terms. In addition, the indexation period for longer dated maturities was lengthened from the previous month to the average of the previous three months.

d. Bank financial operations

In aggregate, commercial banks had low profits in 1994. Gross profits of commercial banks were SIT 4.5 billion in 1994 or 5 percent of equity. This was an improvement over 1993 when aggregate profits were slightly negative. As banks under rehabilitation account for about 40 percent of total assets, their improved financial performance had a large effect on the total.

The main reason for the low profitability in 1994 was the small level of net interest income in 1994 equivalent to only 2 ¾ percent of total assets. 1/ This apparently resulted from a large share of non-interest earning or low interest earning assets—roughly 40 percent of the total assets. 2/ This explanation is, however, insufficient. If all interest income was attributable to the tolar denominated loans, the average rate of return would be 24 percent compared to reported average lending rates of 38 percent in 1994. This implies that the reported average lending rate—the average between the highest and lowest rates charged—overstates rates actually paid. It is possible that many loans were granted to preferred customers at more competitive interest rates. Additionally, doubtful and problematic loans accounted for 16 percent of assets at end-1994 and only partial interest payments were received. Provisions were made against these loans. Provisioning against these loans amounted to SIT 11.0 billion in 1994 or 1.0 percent of assets. Thus, provisioning was not a major factor for the reported wide spreads.

Another explanation is a currency mismatch of assets and liabilities. As of end-1994 foreign exchange assets were 139 percent of foreign exchange liabilities, while tolar assets were only 69 percent of tolar liabilities. 3/ Thus, although tolar deposits were approximately equal to tolar loans, there was a large currency mismatch of other assets and liabilities. This is due to the positive net foreign assets of commercial banks and the high share of foreign exchange denominated Government securities in the portfolios of the banks.

e. Bank Rehabilitation Agency

Progress has been made in resolving the financial difficulties of several problem banks. Three banks, that had originally accounted for about two thirds of banking system assets, were declared insolvent during 1992–93: Lubljanska Banka (LB), Kreditna Banka Maribor (KBM), and Komercialna Banka Nova Gorica (KBNG). A Government study indicated that several factors were responsible for these banks’ poor financial condition: (i) non-performing loans to Slovenian enterprises; (ii) non-performing loans to enterprises in other Republics of the former Yugoslavia (primarily through branches in those republics); (iii) foreign currency deposits whose proceeds were deposited at the National Bank of Yugoslavia (NBY) ; and (iv) excessive operational costs due to over-staffing and inefficiency. After providing emergency liquidity, the Government established the Bank Rehabilitation Agency (BRA) with a mandate to restructure these banks, return them to solvency, and eventually privatize them. LB and KBM were brought under the care of the BRA during 1993 and KBNG in 1994 (after alternative arrangements fell through).

Prior to the BRA take-over of these banks, the BOS initiated several actions to keep the banks afloat. The BOS created special refinancing facilities exclusively to provide liquidity to these banks. In total the BOS supplied credit to LB and KBM equivalent to 3.2 percent of GDP in 1992 and an additional 1.2 percent of GDP during the first half of 1993. In addition, the BOS gave these banks partial exemption from reserve requirements in April 1993.

The BRA initially issued Government bonds worth 12.2 percent of GDP in January 1993 in exchange for non-performing loans of Slovenian enterprises and for claims on the NBY. These 30 year bonds were DM-linked and were not marketable for 5 years. At the same time, the BRA acquired non-performing loans of approximately 100 Slovenian enterprises and set up its own workout unit to collect the outstanding loans either through rescheduling or liquidation of the enterprises. About half the affected enterprises were in the process of liquidation as of mid-1995. At least partial payments were being made by the others. The BRA also replaced the top management of the banks and the new managements initiated restructuring programs in the banks.

In July 1994, the BRA split both LB and KBM into an old bank and a new bank. Old LB and old KBM consisted entirely of assets and liabilities associated with branch operations in other republics of the former Yugoslavia and other financial relations with enterprises in these republics. Assets exceed liabilities in the balance sheets of old LB and old KBMN. Their assets, however, consist of presently uncollectible loans to enterprises in other republics, while their deposits liabilities are to residents of these same republics. The Government considers the financial obligations of old LB and old KBM subject to inter-governmental negotiation as part of the separation agreements with the former Yugoslavia. The other major structural change was a merger between the new KBM and KBNG in January 1995.

The two new banks—LB and KBM—were marginally profitable in 1994—a major improvement over 1993—and now complied with reserve requirement regulations (LB in December 1994 and the merged KBM in May 1995). Liquidity loans to these banks were halved in April 1995 to SIT 7 billion. At end-1994, the new banks exceeded the capital adequacy ratio requirement. The still low level of profits reflects currency mismatch between assets and liabilities, remaining operational inefficiencies, and continued need to provision against non-performing loans. The long position in foreign exchange was due in large measure to the unmarketable Government bonds issued by the BRA which accounted for about 30 percent of LB’s assets and 40 percent of KBM’s assets. The Government’s intention to issue redenominated tolar bonds in exchange for these bonds will provide a vehicle to partially alleviate this source of losses. The BRA intends to put the banks on a sound financial footing before initiating privatization procedures.

f. Bank supervision

The BOS established an active Bank Supervision Department that closely monitored banking activity and vigorously enforced prudential regulations. All banks other than the old LB and old KBM were in compliance with all existing regulations in mid-1995.

The minimum capital regulations will require DM 51 million for a fully operating bank license effective September 1995. This is double the previous minimum capital requirement, which was also doubled in September 1994. The objective of the increases was to force bank mergers. Instead, banks have managed to increase their capital and no mergers have taken place. Effective August 1, 1994, the BOS increased the capital adequacy requirement from 6 ¼ percent to 8 percent on risk weighted assets (in accordance with Basel standards). On average the capital adequacy ratio was 17 percent in mid-1995, with the lowest still 9 percent. Banks, therefore, appear over-capitalized, which in part could account for their low levels of profitability.

The regulations covering provisioning also follows Basel standards and are fully complied with. At mid-1995, provisions equaled SIT 90 billion (about 6 percent of total assets), slightly above the minimum level required on non-performing assets. 1/ The maximum credit exposure to a single borrower or group was lowered to 25 percent from 30 percent of equity capital effective August 1, 1994. The maximum ratio of capital investment plus fixed assets to equity capital was lowered from 1.0 to 0.6. Other prudential regulations cover licenses, accounting standards and documentation, and external auditing. Although no new licenses were issued in 1994, several were issued in 1993. 2/

Banks and savings institution are required to supply data for off-site inspections. A full scope inspection of all banks and savings institutions was in process or had been completed by mid-1995. Once these are completed the BOS plans to initiate a program of bi-annual full scope on-site inspections of each bank and savings institution. The BOS has considerable powers to enforce compliance with prudential regulations and reserve requirements.

g. Government stock operations

The bond market consists only of a very limited number of Government bonds competing with a wider selection of BOS bills. Bonds issued to the rehabilitation banks (DM 1.4 billion) are non-tradeable for 5 years (until 1998). Bonds issued to banks in exchange for claims on NBY (DM 0.9 billion) are non-tradeable for 3 years (until 1996). These two bond issues represent 80 percent of the outstanding government bonds. Four smaller bond issues (RSI, RS2, RS8, and RS11) are traded on the stock exchange (with a total value of DM 0.5 billion). Short-term government securities are traded over-the-counter.

The equity market is very small with only about 50 companies. Most shares are not heavily traded. Market capitalization at end-1993 was DM 220 million in shares and DM 590 million in bonds, of which 90 percent were Government bonds. Market share capitalization rose to DM 460 million in 1994 and for bonds to DM 590 million. Turnover of shares in 1993 was DM 735 million and of bonds DM 807 million. Turnover of shares in 1994 was DM 614 million and of bonds DM 394 million. Trading is dominated by over-the-counter transactions of BOS bills. The decline in turnover during 1994 in comparison to 1993 was attributed to the increase in over-the-counter trading of BOS bills associated with the warrant bills. The stock market is expected to expand following privatization.

2. Ownership transformation. privatization and enterprise reform

a. Introduction

Slovenia is in the process of dismantling the system of social ownership of capital and worker management of enterprises carried over from the SFRY. The absence of a clear distinction between the owners, managers and workers created a situation in which the generation of investible resources and financial viability of an enterprise were subordinated to the objectives of wage escalation and maintenance of inviable levels of employment. The situation was sustained by the access loss-making enterprises frequently had, as owners, to the banking sector. Total accounting losses of enterprises were 19 percent of GDP in 1992. Following an intense debate as to the appropriate methods, a law on ownership transformation was adopted in late 1992 for instituting well-defined ownership and eventual privatization of the socially-owned and mixed enterprises.

b. Legal basis and organizational structures

Slovenia has chosen a hybrid form of privatization which has elements of worker buy-outs and voucher-based mass privatization schemes. The two institutions responsible for supervising, restructuring and assisting the process of privatization are the Agency for Restructuring and Privatization of the Republic of Slovenia (Agency for Privatization, hereafter “Agency”) and the Development Fund (SKLAD). The Bank Rehabilitation Agency (BRA) has also been involved in asset resolution of the loss-making firms in the portfolio of the rehabilitation banks under its auspices.

The Agency monitors and controls, sets guidelines for and approves the privatization programs of socially-owned enterprises (SOEs). SKIAD is responsible for the financial restructuring (to be followed by privatization) of enterprises in its portfolio and is intended to be a temporary depository for the shares of SOEs to be sold to investment funds at auctions organized by it. It also serves as a financing facility for development projects.

The legal basis for the privatization of fully or partially socially-owned companies is the Law on Ownership Transformation (passed in November 1992 and amended in June 1993). Amendments to the law were adopted in 1994 and 1995 to accelerate the privatization process which provide for the Agency to take over the process of ownership transformation of enterprises that failed to submit voluntary privatization program by the deadline of December 1994. They also allow for enterprises to be transferred to SKLAD if the auditing process for them is not completed in a time period specified by the APPNI. Excluded from the law in 1992 were 98 enterprises that were in the portfolio of SKLAD (for details see below) and companies providing special public services (regulated by the Law on Public Service Companies) and banks, insurance companies, enterprises engaged in gambling and those undergoing bankruptcy procedures.

Important supporting legislation included the Company Law, Law on the Securities Market, Law on Mutual Funds and Investment Companies and the Bankruptcy Law.

c. Enterprises under SKLAD

In order to respond more quickly to the problems of especially distressed enterprises, to avoid massive bankruptcies when laws and proper procedures were not well developed, in 1992 the Government invited application, and accepted a total of 98 enterprises to be transformed into commercial companies owned by SKLAD. These firms included some of the largest loss-makers with problems arising from loss of former Yugoslav markets, over indebtedness, and excess labor. These enterprises had 55,000 employees, about 10 percent of business employment, with debt over DM 2 billion and losses at DM 650 million, against their book value of DM 700 million in 1992.

Following legal audits and evaluation of their operational and financial performance, the fundamentally viable enterprises were earmarked for immediate privatization; financial restructuring was essential for those that had serious financial problems and the remainder were candidates for liquidation. By June 1995, SKLAD was still the majority shareholder in 42 enterprises. Of the cases that have been resolved, 12 enterprises have been placed under bankruptcy procedures. A total of 21,000 workers (38 percent of those employed by the 98 enterprises) were released. Losses were reduced from DM 650 million (3.4 percent of GDP) in 1992 to DM 150 million (0.7 percent of GDP) in 1993, and to DM 90 million (0.4 percent of GDP) in 1994.

SKLAD seeks to relinquish ownership or liquidate substantially the remaining enterprises in its portfolio; the sale of an additional 20 enterprises is expected by end-1995. In the meantime, efforts are underway to further restructure these firms by reductions in the labor force, infusion of new financing, debt reduction and restructuring. SKLAD would like to keep these enterprises viable and will undertake limited capital expenditures where it is absolutely necessary (DM 10 million was invested in 1994). Final negotiations are underway for the participation of the EBRD in some of the enterprises.

Under the amendments to the Law on Ownership Transformation two types of enterprises would be eligible for transfer to SKLAD: (i) enterprises which do not complete the audit of their opening balance sheets in the specified time period automatically revert to SKLAD. Sixteen enterprises were received under this category and are intended for liquidation by end-1996; and (ii) enterprises which did not prepare their own autonomous privatization programs by the December 1994 deadline, are first taken over by the Agency and proposed for transfer to SKLAD. About 100 to 150 such cases are expected in 1996.

SKLAD is responsible for organizing auctions of shares of enterprises in its own portfolio and enterprises with second approval of their privatization programs. Three such auctions were held at the Ljubljana Stock Exchange in December 1994, March 1995 and July 1995. In these auctions, SIT 2.7 billion, SIT 11.5 billion and SIT 6.7 billion worth of shares of enterprises, were sold to authorized investment funds. Investment funds can bid for the shares of enterprises with 2 percent increase or decrease at a step of the initial offering price. The initial offering price is based on opening balance sheet of an enterprise. The purchasing price, on average, paid for by investment funds was 94 percent of the initial offering price.

Besides its role in mass privatization and restructuring of enterprises, SKLAD started long-term financing of new projects in 1995 to support private sector in the creation of new employment opportunities. In mid-1995, DM 50 million was available to SKLAD for this purpose. SKLAD is handicapped in assisting large investment projects; it also faces a paucity of good project proposals from the private sector.

d. Socially-owned enterprises

The main objective of the Law on Ownership Transformation is to transform enterprises with socially-owned capital into companies with known owners, chiefly private individuals and institutional owners. The law provides for a combination of free distribution and commercial privatization of shares of these enterprises according to the following formula: 20 percent to the Development Fund for transfer to the investment companies and funds at a later date, 10 percent to the Restitution Fund, 10 percent to the Pension Fund, 20 percent to workers, and the remaining 40 percent for commercial sales (through public offerings of shares, public tender, public auction or sale to insiders on preferential terms). The employees’ shares have been distributed in the forms of vouchers to the population at large, the amount depending upon years of participation in the labor force. The vouchers can either be used to buy shares in the enterprise in which the worker is employed, to buy shares of authorized investment funds or to acquire shares of companies during public bidding. Shares of other firms can not be bought directly. The law allows for the entire or partial sale of the company, with the three funds being compensated from the proceeds of the sale.

A total of 1,400 SOEs, along with their 1,000 daughter companies, with an estimated social capital of US$10 billion fell under the Law on Ownership Transformation. After long delays, the pace of privatization of the SOEs accelerated in 1994. The Agency had received around 1,300 autonomous privatization programs with an estimated social capital of US$9 billion by the deadline of December 31, 1994. About half of these programs were submitted in December alone. By May 1995, about 215 programs with social capital of US$900 million had second approval (meaning end of the privatization program and court registration), 478 programs with social capital of about US$3 billion had first approval but not the second approval, and 686 programs were in procedure for first approval, and 13 programs had been transferred to SKLAD. Companies in good financial status applied for privatization earlier regardless of their size and took less time for second approval. There were about 300 companies, mostly small enterprises which did not submit their autonomous privatization programs. Under the amended laws, the Agency is responsible for carrying out the privatization process for them. A majority of these enterprises are expected to enter bankruptcy or liquidation proceedings.

Internal buyouts were by far the most popular mode of privatization with 90 percent of small- and medium-sized enterprises choosing this option. This method was not viable for large enterprises. These enterprises would necessarily have to resort to a public offering of shares since an internal privatization would be too expensive for the workers and managers. Of the enterprises that submitted privatization programs, a total of 126 enterprises applied for a public offering of their shares. By May 1995, 30 enterprises had completed their public offerings, 25 were in process, and the rest waiting for approval by the Agency. For large enterprises, financial support from the banking sector were necessary. The lack of financing made the private sector’s participation in manufacturing industry difficult.

Investment funds are viewed as a counterbalance to the insider ownership in the privatization process. Privatization investment funds were approved and licensed by the Securities Exchange Agency under the Law on Mutual Funds and Investment Companies (adopted in January, 1994). These funds were allowed to issue equity to acquire ownership certificates from the public which they would, in turn, use to obtain the 20 percent shares of enterprises earmarked for them from SKLAD.

In 1994, twenty three investment management companies were set up to manage about sixty investment funds. The funds are closed-ended; shares of these funds will not be traded on the stock exchange until 1997. The shares of enterprises will be traded only among investment funds, the Pension Fund and the Restitution Fund. By May 1995, the total amount of ownership certificates collected by the investment funds was an estimated SIT 226 billion; 37 funds had completed their collection of ownership certificates. In turn, investment funds had issued shares totalling SIT 229 billion; there is a large variation amongst the funds with some funds having sold all of their shares while others have been able to sell very little.

Investment funds have undertaken a major role in providing information to the public about the opportunities available for investment of ownership certificates. The Government has made efforts to advertise the privatization process abroad and foreigners can enter the process through tender offers or as strategic investors. There is, theoretically, no limit to the amount of foreign participation. However, the decision whether to allow a foreign collaborator rests with the enterprise. Most of the interest for investment in Slovenia firms has come from firms in trading partner countries.

e. Reform of state-owned enterprises

Efforts are under way to convert enterprises in the public service sector into joint stock companies after a determination on the share of state and social ownership has been made. The privatization of the socially-owned component will be implemented under the Law on Ownership Transformation. Besides die law for the railroad sector and the decree on the Transformation of Postal Services (both sectors are to remain largely state-owned) which were adopted by the Parliament in 1993, the Government is at various stages of preparation and processing of laws for the energy, roads, telecommunications, ports and airports, and air and maritime transportation subsectors. The Government also aims to reduce the share of state ownership in the medium term. The Ministry of Economic Affairs has instructed the Agency for Privatization to divest the shares of the state when privatizing the social capital in enterprises with mixed ownership.

The process of consolidation of ownership of state-owned enterprises in the Treasury has been initiated. While the Law on Method and Financing of Transportation of Railway Subsector and on Ownership Transformation of Public Services of Slovenia Railways establish the operating principles for the railroad sector, laws on the other subsectors are expected to be processed and passed in 1996.

f. Enterprises under the BRA

The BRA took over assets involving claims on enterprises in the lowest financial reliability classes from the banks in 1993 and has been managing them under the Instructions on the Treatment of Assets. Among the total 105 companies, 15 companies had debts over DM 10 million each. The BRA tried to sell the assets at a discount, reschedule debt, or convert claims into capital investment in the enterprise for sale at a later date. It aimed at the financial rehabilitation of enterprises and reduction of public debt. The debt of 5 companies was resolved with a reduction in public debt by DM 196 million by May 1995. Forty-five companies were under bankruptcy procedures.

Table 16.

Slovenia: Gross Domestic Product by Sector

(In millions of dinars/tolars. current prices)

article image
Source: Statistical Office of the Republic of Slovenia. Year 1994—estimates of the Institute of Macroeconomic Analysis and Development.
Table 17.

Slovenia: Structure of Gross Domestic Product by Sector

(In percent)

article image
Source: Statistical Office of the Republic of Slovenia. Year 1994—estimates of the Institute of Macroeconomic Analysis and Development.
Table 18.

Slovenia: Cost Structure of Gross Domestic Product

article image
Source: Statistical Office of the Republic of Slovenia; Years 1994—estimates of the Institute of Macroeconomic Analysis and Development.
Table 19.

Slovenia: Supply and Use of Resources

article image
Source: Statistical Office of the Republic of Slovenia; Years 1994—estimates of the Institute of Macroeconomic Analysis and Development.
Table 20.

Slovenia: Expenditure on Gross Domestic Product

article image
Source: Data provided by the Slovenian authorities.

Preliminary estimates by Institute of Macroeconomic Analysis and Development.

Table 21.

Slovenia: Income and Expenditure of Households

article image
Source: Institute of Macroeconomic Analysis and Development.

Preliminary.

Table 22.

Slovenia: Structure of Gross Fixed Investment 1/

article image
Source: 1992, 1993—Statistical Office of the RS. 1994—estimates of Institute of Macroeconomic Analysis and Development.

Includes socially-, state-owned, and private enterprises (with more than three employees).

Table 23.

Slovenia: Investment in Economic Infrastructure

article image
Source: Statistical Office of the Republic of Slovenia—estimates by Institute of Macroeconomic Analysis and Development.
Table 24.

Slovenia: Developments in Production

(Indices: 1990 = 100)

article image
Source: Data provided by the Slovenian authorities.

Real value added.

Table 25.

Slovenia: Industrial Output

(In millions of new dinars/tolars)

article image
Source: Statistical Office of the Republic of Slovenia.
Table 26.

Slovenia: Developments in Tourism

(In thousands)

article image
Source: Statistical Office of the Republic of Slovenia.

Including the former SFRY.

Table 27.

Slovenia: Agricultural Production

article image
Source: Statistical Office of the Republic of Slovenia.
Table 28.

Slovenia: Primary Energy Production and Consumption 1/

article image
Source: Data provided by the Slovenian authorities.

Consumption encompasses all forms of use, including further transformation, additions to stocks, and non-energy uses.

In thousands of terajoules.

Table 29.

Slovenia: Secondary Energy Production and Consumption

(In thousands of terajoules) 1/

article image
Source: Data provided by the Slovenian authorities.

1012 joules.

Primary plus secondary energy imports relative to secondary energy consumption.

Table 30.

Slovenia: Distribution of Enterprises by Sector and Ownership 1990–94

article image
Source: Data provided by the Agency for the Republic of Slovenia for Payments, Supervision and Information (formally SDK).

Includes the numbers of enterprises with mixed ownership.

Table 31.

Slovenia: Distribution of Enterprises by Since in 1994

article image
Source: Data provided by the Agency for the Republic of Slovenia for Payments, Supervision and Information (formally SDK).
Table 32.

Slovenia: Distribution of Enterprises by Ownership in 1994

article image
Source: Data provided by the Agency for the Republic of Slovenia for Payments, Supervision and Information (formally SDK).
Table 33.

Slovenia: Labor Force, Employment, and Unemployment

article image
Source: Data provided by the Slovenian authorities.

Includes socially-, state-owned, and private (with more than three employees). Subsector data are calculated on the basis of march and September data.

End of the year.

Table 34.

Slovenia: Nominal and Real Average Monthly Mat Wage

Per Worker in the Enterprise and Public Sector

article image
Source: Data provided by the Slovenian authorities.

Average net personal income deflated by the consumer price index.

Change from corresponding period of preceding year.

Table 35.

Slovenia: Average Monthly Net Wage, By Sector

article image
Source: Data provided by the Statistical Office of the Republic of Slovenia.

Includes socially, state-owned enterprises and private (with fewer than 3 employees).

Includes forestry and water supply.

Table 36.

Slovenia: Ownership Structure and Employment, 1990-94

article image
Sources: Statistical Office of the Republic of Slovenia; and Register of Enterprises.
Table 37.

Slovenia: Registered Unemployment and Job Vacancies

article image
Sources: Employment Office of Republic of Slovenia; estimates and projections by the Institute of Macroeconomic Analysis and Development.
Table 38.

Slovenia: Output, Employment and Unit Labor Costs in Industry

(Annual percent change)

article image
Source: Data provided by the Slovenian authorities.

In constant prices.

Table 39.

Slovenia: Price Developments

(Percent change)

article image
Source: Data provided by the Bank of Slovenia.

Growth over same month of previous year. Annual data shows changes between average levels of current and previous year.

Annual data are given as the average of monthly growth rates within the year.

Table 40.

Slovenia: Coverage of Enterprise Losses in 1989-94

(In millions of diners/tolars)

article image
Source: Income statements, Agency for the Republic of Slovenia for Payments, Supervision and Information (formally SDK).

Due to changes in income statement reporting, all the data items, except loss in current years, are not available.

Includes enterprises that want bankrupt in 1993.

Data from 1992 financial statements.

Table 41.

Slovenia: Balance of Payments. 1991-94

article image
Sources: Bank of Slovenia; and staff estimates.

For 1991, excluding transactions with the former Yugoslav republics; processing is included in services, excluded in merchandise.

On a c.i.f. basis for 1991, f.o.b. thereafter.

Including net errors and omissions.

Table 42.

Slovenia: Direction of Trade. Exports. 1991-94 1/

article image
Sources: Statistical Office of the Republic of Slovenia; and Bank of Slovenia.

For 1991, excluding transactions with the former SFRY; processing is included in services, excluded in merchandise.

Former EFTA countries are included in EU data after 1992.

Table 43.

Slovenia: Composition of Exports, f.o.b., 1989-94 1/

article image
Source: Statistical Office of the Republic of Slovenia.

Trade with other republics of the former SFRY is not included prior to 1992. 1992 and 1993 include processing.

These totals may be slightly different from those of the balance of payments.

Table 44.

Slovenia: Direction of Trade, Imports. 1991-94 1/

article image
Sources: Statistical Office of the Republic of Slovenia; and the Bank of Slovenia.

On a c.i.f. basis.

Former EFTA countries are included in EU data, after 1992.

Table 45.

Slovenia: Composition of Imports, c.i.f.. 1989-94 1/

article image
Source: Statistical Office of the Republic of Slovenia.

Trade with other republics of the former SFRY is not included prior to 1992.

These totals may be slightly different from those of the balance of payments.

Table 46.

Slovenia: Services Balance, 1992–94

(In million, of U.S. dollar.)

article image
Source: Data provided by the Bank of Slovenia.

The BOP item “Financial services” in this table differs from Financial Services published in Monthly Bulletin Table 2.3.

Table 47.

Slovenia: External Debt. 1990–94

article image
Source: Data provided by the Bank of Slovenia.

Excluding non-allocated debt and the IMF.

Including liabilities of the NBY for loans transferred to Slovenian final beneficiaries.

Arrears of principal due and not paid in the first stand-still period (July 1, 1989-June 30, 1990) and related to original i.e., non-rescheduled credits extended or guaranteed by the export credit agencies; obligations included under the item: Private non-guaranteed debt.

Including interest arrears on long-term debt in the amount of US$22 million.

Table 48.

Slovenia: External Debt Service, Long-Term Debt. 1992–94 1/

article image
Source: Data provided by the Bank of Slovenia.

Excluding non-allocated debt and the IMF.

Payments of: (i) principal and interest under the Yugoslavia New Financing Agreement, 1988 (debt restructured with commercial banks) have for defined Slovenian obligation thereunder, together with (ii) 16.3 percent of interest due on the obligation of the National Bank of Yugoslavia under the said agreement; and (iii) principal of and interest on deposit with the former Ljubljanska banks d.d. Ljubljana made by the National Bank of Yugoslavia using funds deriving from the Yugoslavia Trade and Deposit Facility Agreement, 1988, have—according to the Government’s decision of January 13, 1994, in 1994 been made to the fiduciary account of the BOS with the Dresder Bank Luxembourg St., Luxembourg, They are not shown in the table above.

Table 49.

Slovenia: Monetary Survey, 1991–95

article image
Source: Data provided by the Bank of Slovenia.
Table 50.

Slovenia: Monetary Authorities (Bank of Slovenia), 1991–95

(In billions of tolars)

article image
Source: Data provided by the Bank of Slovenia.
Table 51.

Slovenia: Deposit Money Banks Balance Sheet, 1991–95

(In billions of tolars)

article image
Source: Data provided by the Bank of Slovenia.
Table 52.

Slovenia: Selected Interest Rates. 1992–95

(Annualized, in percent)

article image
Source: Data provided by the Bank of Slovenia.
Table 53.

Slovenia: Summary of General Government Operations, 1992–95

article image
Source: Ministry of Finance of the Republic of Slovenia.

Transfers between the different levels of government and the funds are netted out.

Consisting of unemployment and maternity contributions; and, during 1991, health contributions.

Table 54.

Slovenia: General Government Revenue. 1992–95

(in billion of SIT)

article image
Source: Ministry of Finance of the Republic of Slovenia.
Table 55.

Slovenia: General Government Revenue, 1992–95

(In percent of GDP)

article image
Source: Ministry of Finance of the Republic of Slovenia.
Table 56.

Slovenia: General Government Expenditure, 1992–95

(In billions of SIT)

article image
Source: Ministry of Finance of the Republic of Slovenia.

Includes health expenditures in 1991.

Includes transfers to the federal budget in 1991 of SIT 3.04 billion.

Table 57.

Slovenia: General Government Expenditure, 1992–95 1/

(In percent of GDP)

article image
Source: Ministry of Finance of the Republic of Slovenia.

Includes central government, Pension Fund, and Health Fund, excludes localities.

Includes transfers to the federal budget in 1991 of SIT 3.04 billion.

Table 58.

SLOVENIA: and the EU: A Comparison of Revenues and Expenditures 1/

(As percent of GDP)

article image
Sources: OECD (1995). OECD (1994); and Slovenian Ministry of Finance.

The data on expenditures and revenues for the EC countries are for 1993; the tax structure data is for 1992. Data for Slovenia refers to 1994.

ANNEX: The External Trade and Payments System 1/

1. Exchange Rate Arrangement

The currency of Slovenia is the tolar (SIT). 2/ The exchange rate for the tolar is determined freely in the exchange market by the interplay of supply and demand. The Bank of Slovenia may also buy and sell foreign exchange in transactions with authorized banks and the Government.

The buying and selling rates for cash transactions in exchange offices in deutsche mark, Austrian schilling and Italian lira are published every day by most newspapers. Cross rate quotations are determined by market forces. For transactions between companies and banks, rates are published every day by the bigger banks, mostly in deutsche mark. For June 1994, the banks’ middle rate was SIT 79.59 per DM. The BOS does not prescribe a spread between buying and selling rates for banks’ dealings with the public. Banks are free to set commissions. For the month of June, 1994 the official exchange rate was SIT 129.92 per US$. There are no taxes and subsidies on foreign exchange transactions.

Natural persons conduct foreign exchange transactions in currency notes and travelers’ checks, usually with foreign exchange offices, at freely negotiated rates. Companies repatriating foreign exchange have two business days to sell their proceeds to other companies at freely negotiated exchange rates or to use them for payments abroad. After this period, they must sell these proceeds to authorized banks which, in turn, may sell foreign exchange to companies for payments and transfers abroad.

The BOS publishes daily a moving two-month average exchange rate for customs valuation and accounting purposes. The BOS has stated that it intends that this exchange rate not diverge by more than 2 percent from actual exchange rates prevailing in the market. There is no forward market in foreign exchange, but since July 1992 forward foreign exchange transactions are permitted.

2. Administration of Control

The Parliament has authority to adopt legislation in the exchange and trade control area. The Law on Foreign Currency Business and the Law on Credit Transactions with Foreign Countries, adopted mid-1991, are the basic legislation in the exchange control area. Under this legislation, the BOS is delegated regulatory powers in a number of areas, (e.g., the approval of financial credits to nonresidents). In particular, the BOS can limit external borrowing and lending in light of balance of payments needs. The Executive Council (Government) can also pass such measures when measures by the BOS fail to redress the situation, but only for a period of six months and not for transfers abroad of profits from capital investments by foreigners in Slovenia or personal income and indemnities paid to foreign persons in Slovenia. The Lew on the BOS gives the central bank full authority over the formulation and execution of exchange rate policy, including the adoption of exchange rate arrangement. The BOS issues licenses allowing commercial banks to engage in foreign exchange transactions. Credit transactions with nonresidents must be registered with the BOS, which may also grant permission for residents to open bank accounts abroad.

Foreign trade policy is governed by the Law on Foreign Trade Transactions (adopted by the Parliament on March 27, 1993) and is the jurisdiction of the Ministry of Economic Relations and Development. The Law on Customs Service regulates the organization and operations of the customs administration, which is part of the Ministry of Finance. Import quotas are established by the Executive Council and administered by the Slovenian Chamber of Commerce. The contractual joint venture contract is to be submitted to the Ministry of Economic Relations and Development. In order to acquire legal status, equity joint ventures must be entered in the Local Court Register.

3. Prescription of Currency

Slovenia residents may make and receive payments and transfers in any convertible currency. The currencies of Croatia, the former Yugoslav Republic of Macedonia (FYRM), and Bosnia-Herzegovina are considered nonconvertible. Residents are free to agree on settlements in convertible currencies with residents of these countries. Slovenia maintains bilateral payment agreements with Bosnia-Herzegovina (not yet ratified by Bosnia-Herzegovina Parliament) pursuant to which settlements between residents of each country may be made through nonresident accounts with commercial banks. The bilateral payment agreement with Croatia was cancelled. A bilateral payment agreement with FYRM is also maintained under which payments may be made through nonresident accounts with domestic banks or through clearing accounts (with Ljubljanska Banka and Rational Bank of Macedonia). These clearing accounts are not operational.

Under two 1955 agreements (Trieste and Gorica), payments between residents of Slovenia and Italy residing within 20 km of the two countries’ common border may be made through a bilateral autonomous compensation accounts. On the Italian side, both accounts are held at Banca d’Italia while on the Slovenian side, they are held at the Bank of Slovenia and at Ljubljanska Banka d.d., Ljubljana (a private commercial bank), respectively.

The use of foreign currency clauses in contracts between residents is permitted. However, all contracts between residents must be settled in domestic currency.

4. Resident and Nonresident Accounts

Resident natural persons may open foreign exchange accounts in Slovenia. Starting April 27, 1991, banks were required to register new foreign exchange deposits separately and to service new deposits in full. Banks must maintain on deposit abroad a fraction (ranging from 5 to 100 percent depending upon the maturity of domestically held assets) of any net increase in sight foreign exchange deposits beyond the April 27, 1991 level. They must also hold abroad an amount equivalent to 35 percent of their average monthly turnover of convertible payments abroad over the last three months. Resident natural persons may not maintain foreign exchange accounts abroad or hold other financial assets abroad. Foreign exchange accounts dating prior to April 1991 have de facto been frozen. The problem of these old deposits is being solved by banks spontaneously as their financial situation permits but will have to be serviced before the end of 1995.

Resident legal persons and private entrepreneurs (“resident companies”) may maintain foreign exchange accounts in Slovenia only for purposes laid down by the Law, after which no specific authorization by the BOS is required. They may hold such accounts abroad only in special cases (enterprises that perform services in international transport and insurance, representation abroad and payment of taxes abroad), with the permission of the BOS and subject to a reporting requirement. With permission of the BOS and subject to a reporting requirement, but not to any specific conditions, such as use for specific purposes, resident companies may open local currency accounts in Bosnia-Herzegovina, Croatia, and the FYRM.

Nonresidents may open foreign exchange accounts with authorized banks in Slovenia. These accounts may be credited freely with foreign exchange and debited for payments abroad or for buying tolars.

Nonresidents may open local currency accounts with the proceeds of the sale of goods and services to residents or foreign exchange. Balances in these accounts may be used freely to buy goods and services in Slovenia. Balances on these accounts are fully convertible and may be repatriated abroad freely.

5. Imports and Import Payments

Imports from the Federal Republic of Yugoslavia (Serbia and Montenegro) are banned in accordance with a U.N. Security Council resolution. The imports of certain textiles and agricultural products (most live animals and meats, milk and some dairy products) are subject to quota restrictions. The items subject to quota account for 3.2 percent (out of a total of 7,757 tariff items). A few other imports (arms, ammunition, illegal drugs) are banned unless specifically allowed under license.

The Customs Tariff Law of the former SFRY continues to apply (the new Law was passed in January 1995 and will come into force in January 1996) with customs duty rates ranging from 0-25 percent (an average of 13 percent, excluding exemptions). Proceeds are used for insurance and export promotion. Most favored nation treatment is given to all countries. An equalization surcharge of 1 percent and a 1 percent customs fee apply to all imports. Slovenia has an Agreement on Cooperation (establishing a free trade area) with the FYRM since 1992 and signed a free trade agreement with the Czech Republic and Slovak Republic in 1993 the main provisions of which relate to the scope and schedule of tariff reductions on industrial products over a period of two years.

A uniform tariff rate of 15 percent may be charged for goods imported by individuals, goods intended for use by legal entities, associations and other organizations if the value of the imported goods does not exceed tolar 200,000. However, a request for that a normal custom procedure be applied can be made. Customs tariffs are then assessed in accordance with the Law on Customs Tariff. The exemption limit for articles for personal use brought by citizens returning from travel abroad is the total equivalent of US$100 (exclusive of tobacco products, wine and perfume); for citizens who have been working abroad for at least 24 months in the preceding four years the limit (excluding cars) is the tolar equivalent of US$10,000 (US$15,000 for business inventories). In addition, citizens working abroad are allowed a yearly exemption from import duties of US$700.

Foreign exchange for import payments may be purchased freely by companies. Natural persons may purchase foreign currency notes from authorized banks and foreign exchange offices or use funds deposited in their foreign exchange accounts after April 27, 1991 to make import payments.

6. Payments for Invisibles

Resident legal entities may purchase foreign exchange freely to pay for invisibles subject to the same documentation regulations as merchandise imports. Natural persons may purchase foreign exchange from authorized banks and foreign exchange offices or use funds deposited in their foreign exchange accounts after April 27, 1991 to make such payments. Supporting documents on the nature of the services bought abroad or transfers made, such as invoices and health declarations need to be provided. Domestic and foreign currency notes up to the equivalent of DM 1,000 may be taken abroad. Travelers’ checks can be bought from authorized banks under the same conditions as payments for invisibles.

7. Exports and Export Proceeds

Exports to the Federal Republic of Yugoslavia (Serbia and Montenegro) are banned under relevant U.N. Security Council resolutions. Except for a short list of items banned for security and public health reasons, all exports are free. Exports of textile products to the European Union (EU) are governed by the Agreement between the EU and the Republic of Slovenia Trade in Textiles and Products; textiles originating in Slovenia are free from quantitative limits.

Exporters are free to agree to payment terms with foreign importers. However, if the collection of export proceeds is delayed by more than one year, the transaction must be registered with the BOS. Once received, export proceeds must be repatriated. Exporters have two business days to sell their proceeds to importers at a freely negotiated exchange rate or to use the proceeds for payments abroad. After that term, they must sell their proceeds to an authorized bank.

Exporters are exempt from custom duties payable on raw materials and intermediate goods and semi-processed goods used to produce export goods, provided that the value of the exports is at least 30 percent higher than the value of imports and there is no domestic production of the imported goods.

8. Proceeds from Invisibles

No exchange control requirements attach to proceeds from invisibles.

9. Capital

Borrowing and lending operations with foreign countries are not allowed for resident natural persons. Resident juridical persons may borrow abroad on commercial terms without restriction. Companies may raise financial credits abroad but must offer the foreign exchange proceeds to the BOS. The Government may, however, issue guarantees and borrow abroad only in accordance with specific laws passed by Parliament. Short-term bank borrowing abroad is restricted to 15 percent of the minimum amount of the foreign exchange that they are obliged to hold on certain accounts abroad and in the country since September 1994. The Republic of Slovenia can contract or guarantee foreign loans only on the basis of a special law passed by the Parliament that identifies the sources of repayment. Companies may give financial credits only to their foreign subsidiaries with means generated by foreign profits. Banks may extend financial credits to a foreign entity provided that they receive insurance.

Borrowing and lending operations with foreign countries become valid only after registration with the BOS. Except for international refinancing or rescheduling agreements, the early repayment of registered claims and liabilities is permitted.

The Law on Foreign Investment for the former SFRY is still applicable though a new law is under consideration. Direct foreign investment is allowed with only a few exceptions (arms production, part of telecommunications). Foreign investors are eligible for a three year tax holiday from the corporate income tax and may import capital goods required for the investment duty free. There is no restriction on the transfer abroad of interest, profits and dividends from capital investments and construction works projects in the country, provided all obligations within the country have been met. With the concurrence of foreign governments, the bilateral investment protection agreements concluded by the former SFRY continue to remain valid for direct investment in Slovenia. Outward direct investment requires approval from and registration with the Ministry of Finance. Such approval is usually given for the initial investment. Subsequent investment must be financed from either profits (reinvestment) or loans received abroad.

Except for inheritance, nonresident natural persons not performing economic activity in Slovenia may not acquire real estate in Slovenia. Nonresident companies may buy real estate as part of their registered direct investment. Convertibility and transfer of inheritance is allowed under the condition of reciprocity only.

APPENDIX I: Inflation, Exchange Rate, and Demand for Money During the Transition, 1991–95

l. Introduction

Slovenia offers an interesting study in monetary and exchange rate management of a transitional economy. In the approximately 3 ½ years since monetary independence, inflation has declined from near hyper-inflationary levels—20 percent per month or an annualized rate of nearly 900 percent—to near-EU levels. To explain this success, this note describes Slovenia’s monetary policy and examines empirical estimates of money demand.

Use of the exchange rate as a nominal anchor was initially considered, but was not an option because of the Bank of Slovenia’s low international reserves. 1/ The institutional framework, however, was also unfavorable for using money supply as a nominal anchor due to the high degree of wage indexation, financial asset indexation, and a high share (nearly 50 percent) of foreign exchange denominated deposits in broad money. Tolar deposits were indexed to previous month’s inflation and longer-term deposits received a positive real return. 2/ Thus broad money was predominantly indexed percent) either to inflation or the exchange rate, making it highly endogenous. Several authors have hypothesized that the endogeneity of broad money renders money-based stabilization programs less effective (e.g., Calvo and Vegh (1994) and Hoffmaister and Vegh (1995)). Nevertheless, Slovenia was able to overcome these obstacles and bring inflation under control with a primarily money-based approach. Another intriguing aspects of Slovenia’s anti-inflation efforts was that even with a money-based stabilization program, a strong output recovery took place. 3/

Slovenia’s money-based stabilization program concentrated on reserve money and narrow money (M1), the only predominantly unindexed monetary aggregates. 1/ Empirical estimates in this study (Section 4) indicate that the demand for real narrow money was a stable function of real income, real tolar interest rates, the exchange rate, and a positive time trend owing to a remonetization throughout the period under consideration. In contrast, the demand function for tolar time deposits (and M2) was found to be volatile and unpredictable. The estimates of the demand for foreign exchange deposits (and M3) proved to be a stable function. But as broad money in tolar terms was largely endogenous, this broad money supply was not a suitable nominal anchor. Efforts to lower inflation by managing broad money could be frustrated by exchange rate developments with the high share of foreign exchange denominated or indexed deposits and the backward-indexation of tolar assets to inflation.

Initial confidence in the new currency was very low in October 1991, leading to a sharp drop in real money demand and the exchange rate. By halting growth in reserve and narrow money, particularly in the first few months following the introduction of the tolar, the authorities succeeded in establishing a nominal anchor to their financial system. In turn, unexpected external surpluses were created, owing in large part to an undervalued exchange rate induced by low confidence. Later, rising industrial production reinforced confidence in the sustainability of macroeconomic policies. The demand for various monetary aggregates rose rapidly throughout the period mid-1992 to mid-1995. The authorities supported output recovery by accommodating this increased demand for money and following a real exchange rate rule to pursue external competitiveness.

The real exchange rate rule, however, worked against the inflation objective and little progress was made in reducing monthly core inflation from mid-1992 to end-1994. Inflation only fell to a single-digit annual rate in 1995, following a period of stability of the nominal exchange rate in the second half of 1994 and the first half of 1995. Thus, as has been experienced in other countries (Calvo and Vegh (1994)), the money-based stabilization approach encountered an apparent threshold in its effectiveness without further support from the exchange rate or incomes policy. Granger causality tests (Section 3) confirm that in Slovenia inflation was strongly “caused” by exchange rate movements and only weakly “caused” by Ml. The exchange rate movements were found to be caused by reserve money expansion and vice versa, although much weakly. This suggests that monetary policy had an independent impact on the exchange rate but the channel by which inflation was lowered worked primarily through the exchange rate. This is not a surprising conclusion given that Slovenia is a highly open and indexed economy.

This note presents estimates of the causality between monetary, inflation, and exchange rate variables, and econometric estimates of demand for money. Although the extremely short time period since monetary independence (and hence the reliance on high frequency data) hinders empirical analysis, the results are suggestive and provide an insight into the reasons for Slovenia’s monetary policy successes. Section 2 describes monetary policy targets, institutional background, and reviews the outturn during 1991–95. Section 3 discusses causality findings. Section 4 is an empirical examination of demand for money. Section 5 offers some conclusions.

2. Monetary policy

In 1991, the authorities were confronted with an extremely difficult situation. Inflation was high and the economy was highly indexed. International reserves were virtually non-existent and confidence in the currency and the level of the exchange rate was meager. The authorities did not have the international resources to use the exchange rate as a nominal anchor and even if they did, it risked an adverse output and external response owing to backward-indexation. A money-based approach also posed complications. Most monetary aggregates were highly endogenous because of the high proportion of foreign currency deposits and backward indexation of tolar deposits. Moreover, existence of a stable/predictable money demand function could only be postulated for this new economy/country undergoing structural transformation. The immediate goal of monetary policy following monetary independence was to eliminate the double-digit monthly inflation (over 20 percent in October 1991) that was inherited from the former Yugoslavia. The BOS endeavored to establish its credibility by distancing its policies from those of the National Bank of Yugoslavia (NBY).

Initially, faced with this uncertainty, the BOS induced a nominal decline in reserve money during October 1991 to March 1992, which produced a real decline of 55 percent, and a one-third real decline in narrow money. The BOS also established a floating exchange rate regime. The exchange rate depreciated substantially, falling to one-fourth of its value over October 1991-January 1992. The nominal exchange rate subsequently stabilized during February-June 1992. During April-June 1992 the BOS allowed reserve money and narrow money to grow to accommodate increases in real money demand. Inflation fell to 2–3 percent per month by mid-1992. An overshooting of the real exchange rate is evident as a current account surplus of 7 ½ percent of GDP was recorded in 1992.

Beginning mid-1992, the BOS settled into a new monetary stance. Its goal was to gradually reduce inflation, while assuring a quick recovery for the real economy. The monetary authorities rejected a fixed exchange rate as a nominal anchor. It was feared that the real exchange rate would appreciate too much and too fast, harming export competitiveness and delaying recovery (Mencinger (1994) and Bole (1995)). The authorities were unwilling to use the exchange rate to reduce inflation if a real currency appreciation that were to forestall an output recovery.

The BOS, noting there were two major goals—lowering inflation and stimulating output—relied upon two tools to achieve its targets. The first tool was a reserve money program, which was primarily oriented toward the inflation target. At the same time to avoid an overly restrictive money stance that would stifle economic growth, explicit adjustors for growth in real income (and transactions) were included in the reserve money program. The second tool was a real exchange rate rule designed to preserve export competitiveness and promote real GDP growth as exports accounted for 60 percent of GDP. This approach required substantial sterilized intervention in the foreign exchange market. To limit losses from sterilization, the BOS used refinancing tools and prudential regulations to induce banks to acquire foreign assets.

The reserve money program was characterized by inflation targeting. Each year, unannounced monthly inflation targets were chosen. The monthly growth of reserve money was set consistent with desired inflation. However, since the BOS was uncertain about the money demand function, several factors were employed to adjust the base line path for reserve money. Even though later econometric estimates would indicate a stable money demand function, at the time the data series was too short to yield precise estimates (Bole (1995)). The first adjustment factor was the direction in last month’s inflation from the inflation target. Reserve money growth was reduced by a proportion of the excess of actual inflation over the target level. The second factor was real household income (available with two months lag) and the third factor was real transactions (available with three months lag). Faster growth of the last two variables signalled a higher transaction demand for money that needed to be accommodated to protect output recovery. The target was also adjusted for seasonal and other intuitive factors. The policy could thus be represented by the following feedback equation:

dlogRM(t) = dlogP* + α [ dlogP* - dlogP(t-1)] + β dllogHH/P(t-2) + γ dlogTrans/P(t-3)
article image

This reaction function could produce a negative correlation between prices and money as lower inflation would lead to faster reserve money increases and the reverse.

Although inflation subsided from mid-1992 to end-1992 (1 ½ percent per month), from end-1992 to end-1994 core inflation was essentially stuck. Declines in administered price adjustments lowered total inflation gradually during 1993–94. 1/ Thus, although it may have appeared initially that there were two separate instruments to achieve two goals, it was discovered that these tools were not sufficiently independent. The real exchange rate rule worked against the inflation reduction and the monetary program was not restrictive enough to overcome this channel of inflation inertia. Later with support from the incomes policy in 1994, the authorities gradually replaced the real exchange rate with a greater emphasis on exchange rate stability, which succeeded in lowering inflation sharply in 1995.

3. Determinants of inflation

To gain insight into the effects of monetary policy on inflation, Granger causality tests were conducted on several variables (Table 59). Tests of causality between the exchange rate and, either core inflation or total inflation, indicate a bidirectional causality with more significant results obtained for core inflation and the exchange rate, as might be expected. The time lag is very short, only two months. The exchange rate is found to have a significant and positive effect on core inflation. Core inflation has a significant positive one period lag impact and a negative two period lag effect. This bidirectional causality between the exchange rate and core inflation is not surprising given the openness of the economy and the existence of a real exchange rate rule for much of the sample period.

Table 59.

Slovenia: Granger Causality Estimates 1/

article image
article image
article image
article image

Statistical evidence indicates that Ml has a weaker impact on total inflation and with a longer lag than the exchange rate. In a test using a six-period lag structure, all the lags are positive, but only the sixth lag is statistically significant and a joint test of significance of the fourth through sixth lag is only weakly significant at the 90 percent level of confidence. A Granger causality test in the reverse direction—total inflation on M1—shows alternating signs and a statistically significant positive effect of the sixth lag (99 percent). The Granger causality tests were insignificant in both directions for M1 and core inflation. Tests of Granger causality of M2, tolar time deposits, foreign exchange deposits, broad money, and on either inflation or core inflation failed to produce significant results in either direction.

Finally, tests indicated bidirectional causality between reserve money and the exchange rate. The effect of reserve money on the exchange rate was stronger with a lag of 3 to 6 months and was statistically significant at the 90 percent level of confidence. For the exchange rate on reserve money, only the second period lag is positive and statistically significant.

4. Econometric estimates of demand for money

Despite the short time series available and high frequency data (about 40 monthly observations from January 1992 to May 1995), empirical estimates were undertaken to shed light on events. While long term trends cannot be expected to be identified with such a short sample period, particularly with several major policy regime changes, the findings nevertheless indicate that real narrow money and broad money are stable functions.

An error correction model (ECM) was estimated for narrow money (M1), tolar time deposit M2, foreign exchange deposits and M3. The theoretical model posits the long-term relationship as follows:

log ( M1 ) = alog(P) + (1-a) log (E) + blog(HH/P) - c(r),
article image

In testing the model, the coefficient on real household income obtained was not significantly different from unity. Subsequently, with the coefficient on income (b) constrained to be unity, this equation can be transformed as,

log ( M 1 / HH ) = - ( 1 - a ) log ( P/E ) - c ( r ) ,

which is the basis for the final form of the ECM that was tested. In this formulation, the exchange rate is a proxy for inflationary expectations. Purchasing power of tolar M1 holdings are evaluated effectively both in terms of tolars and DM equivalents.

Several different formulations were tested and insignificant variables omitted. The empirical results reported herein are estimates of the separate components of broad money. Estimates of equations for M2 and broad money were made for purposes of comparison, but are not reported since they added little. The estimated equations used the following as independent variables: the exchange rate, real interest on tolar deposits, and real interest on foreign exchange deposits. These represent the separate components that influence the rates of return to different types of money. The calculated actual rates of return to Ml, tolar time deposits, and foreign exchange deposit were found to be insignificant. The probable reason is that the separate components had very different patterns of adjustment. Finally, coefficients on inflation and the interest rate on foreign exchange deposits were found to be insignificant and dropped for simplicity. 1/ Additionally, transactions and industrial production variables did not prove significant.

The estimated equation for M1 consists of first difference variables and the error correction term. The dependent variable is the change in the log of M1 (dlog(M1)). The independent variables are change in the log of real household income (dlog(HH/P)), change in the real interest rate on tolar deposits (dr), change in the log of the tolar-DM exchange rate (dlog(E)), a trend variable (T), and monthly seasonal dummy variables. The error correction term is a coefficient on lagged variables (one period) of the log of velocity (log(M1/HH)), the log of prices divided by the exchange rate (log(P/E)), and the real tolar deposit rate (r). The first two level terms can be transformed into log(M1), log(HH/P), log(M1/P) and log(M1/E) as shown above. The results based on OLS estimations are as follows (t-statistics are in parentheses below the variables; see Table 60 for diagnostics).

Table 60.

Slovenia: Demand for M1

Dependent variable: dlog(M1)

article image
article image

January dummy variable; December is the default.

d log ( M1 ) = - 3.3 ( 3.9 ) * * + 0.52 ( 5.7 ) * * ( d log ( HH/P ) ) - 0.052 ( 3.6 ) * * ( dr ) + 0.46 ( 2.8 ) * ( d log ( E ) ) + - 0.59 ( 4.2 ) * * log ( M1/HH ) - 0.38 ( 3.1 ) * ( log ( P/E ) ) - 0.045 r ( 5.1 ) * * + 0.0074 ( 3.6 ) * * T  + seasonalvariables .

R-squared: 0.944; F(18, 21)= 19.7; DW = 2.05.

** 99 percent level of confidence

* 95 percent level of confidence

The implied long-term solution to the model is,

- 0.59 [ log ( ml ) - 0.36 log ( P ) - 0.64 log ( E ) - log ( HH/P ) + 0.076 r 0 .0125T] .

The results imply that M1 is a stable function of the above variables. The error correction term coefficient of 0.59 is quite high and suggests a rapid (3 month) adjustment to long-run values. The influence of both the domestic price level and the exchange rate on M1 is strongly borne out (Chart 15).

CHART 15
CHART 15

SLOVENIA: REGRESSION ESTIMATES FOR M1

(In billions of tolars)

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by Slovenian authorities.

The short-term effects are indicated by the coefficients on the change in log(HH/P), change in the real tolar interest rate, and the change in the exchange rate. All have the expected sign. Increases in real income increase the transaction demand for M1. Increases in real interest rates have a negative effect, as expected, because the opportunity costs of M1 rises. Demand for money increases with the exchange rate, apparently because of increased transactions demand. The time trend is significant and positive, suggesting a remonetization during this period.

a. Tolar time deposits

The equations for tolar time deposits was set-up as an error correction model using a similar equation, except for the treatment of the exchange rate (Table 61). None of the coefficients, however, were significant except the time trend and the error correction term on real tolar time deposits. The low DW statistic (1.36) and an R-squared of 0.64 indicate that a great deal is unexplained. Thus, the variables included in the analysis do not provide insight into the behavior of tolar time deposits (nor for M2 for which the empirical results were less significant).

Table 61.

Slovenia: Demand for Tolar Time Deposits

Dependent variable: dlog(TD)

article image
article image

January dummy variable; December is the default.

b. Foreign exchange deposits

The estimation of foreign exchange deposits equation fares much better, (Table 62). 1/ The results imply a well defined function for foreign exchange deposits (and broad money). Prices, exchange rate, and interest rate changes are found to influence foreign exchange deposits. Exchange rates have the expected positive sign and exceed unity since the tolar value of foreign exchange deposits rises automatically with the exchange rate. Furthermore, exchange rate movements affect the tolar rate of return to foreign exchange deposits. Changes in real interest rates on tolar deposits (r) have a negative coefficient as expected since it represents an opportunity cost to holding foreign exchange deposits. However, the coefficient associated with the lagged level of r was not significant. Inflation has a positive coefficient, implying that inflation is associated with an increase in foreign exchange deposits (tolar valued). The time trend coefficient in this case is negative, confirming the tendency towards switching out of foreign assets into tolar assets over this time period.

Table 62.

Slovenia: Demand for Foreign Exchange Deposits

Dependent variable: dlog(FXD)

article image
article image

January dummy variable; December is the default.

With a money-based anchor and the exchange rate market determined, the tolar value of foreign exchange deposits becomes endogenous. Thus stability of this function does not provide a basis for the conduct of monetary policy when such a large portion of broad money is foreign exchange denominated.

5. Conclusion

Slovenia’s money-based stabilization was successful during 1991–95 as is borne out by the steep decline in inflation. On closer examination of the different sub-periods, however, a somewhat more complex picture emerges. A massive depreciation of the currency took place from October 1991 to January 1992, as confidence in the currency and the monetary authorities was virtually non-existent. The Slovenian monetary authorities instituted a major decrease in reserve money from October 1991 to March 1992 to anchor the financial system and expectations. The overshooting of the exchange rate facilitated a current account surplus in 1992, an accumulation of international reserves, and stabilization of the exchange rate during February 1992 to June 1992. The recovery in money demand and the nominal exchange rate, succeeded in reducing inflation to 2 ½ percent per month by mid-1992.

Thereafter, the authorities set out to control inflation with a money-based program with a goal of supporting a recovery of output through the use of a real exchange rate rule. Beginning in mid-1992 the BOS adopted a reserve money program based on implicit inflation targeting. Because of uncertainty about demand for money, the reserve money program was adjusted monthly to avoid too loose or too tight a monetary stance that might choke off the output recovery. Clearly, a stable demand for money function was needed for a monetary anchor to work. With tolar deposits and foreign exchange deposits indexed, M2 and broad money were either unstable (the former) or highly endogenous (the latter) with a real exchange rate rule or a floating exchange rate system in place. The empirical results indicate, however, that the demand for narrow money was stable to provide the nominal anchor to the financial system. Therefore, monetary policy produced successful disinflationary results because it was oriented towards nominal variables—reserve money and M1—that were for the most part not indexed to inflation or the exchange rate.

The BOS wanted both to reduce inflation and to maintain a real exchange rule. However, the real exchange rate rule worked against their disinflationary objective. Core inflation became entrenched at above 1 percent per month. With the passage of an incomes policy designed to limit real wage increases in 1994, an alternative means of protecting external competitiveness was introduced. Thus, the BOS gradually eased the real exchange rate rule, the nominal exchange rate stabilized, and inflation declined to near EU levels. The evidence implies that inflation was reduced most effectively through the exchange rate channel. Indeed, core inflation was not lowered significantly during 1993–94 when the real exchange rate rule prevailed. It appears that in this small open and highly indexed economy, inflation was primarily caused by the exchange rate. Monetary policy was successful to the extent that it achieved a stable nominal exchange rate.

APPENDIX II: The Future of Pensions in Slovenia

The Slovenian pension system is in difficult financial straits. Given current demographic trends, further deterioration in its financial balance is likely over the near-term. Therefore, pension benefits will need to be pared back to contain budget expenditures. In the long run, to reduce payroll taxes (and, ultimately, labor costs), and to encourage savings Slovenia may need to combine its pay-as-you-go (PAYG) system with a fully-funded system.

Most of these assertions—with the exception of the financial situation—require some qualifications. This section discusses these qualifications and attempts to shed light on the complex issues confronting Slovenia.

A central question is the extent to which demographic forces are creating strain on the pension system. The transition to a market economy, together with various micro-inefficiencies in the provision of pensions, are responsible for the current condition of the pension system. Longer life expectancy and lower fertility rates have only begun to exact their toll on the pension system and further deterioration is in the offing. But even when demographic pressures reach their zenith, around 2030, the budgetary costs of pensions could be held in check with only relatively minor changes to benefits, if these changes were implemented soon enough.

In Slovenia, as well as in many other transitional economies, the budgetary cost of pensions has increased markedly since the beginning of the transition. From 1991 to 1994, various costs associated with the social safety net (pensions, health expenditures, and unemployment benefits) have increased by 5 percent of GDP (Table 63). Unemployment benefits have been contained to less than 1 ½ percent of GDP on average during this period; pensions and health expenditures have thus caused the bulk of the increase in social safety net expenditures.

Table 63.

Social Transfers

(As percent of GDP)

article image
Source: SORS, various publications.

Including old-age, disability and survivors.

This rapid deterioration is, in itself, telling. Indeed, it cannot be attributed to a decline in GDP—Slovenia’s output has largely recovered to its pre-transition level—or to increasingly generous pension benefits—they are lower in real terms than in the 1980s.

The transition to a market economy rather than pure demographic forces has been the main cause behind the higher cost of Slovenian pensions. The reasons for this are manifold. First, the government used pensions to shelter certain individuals from some undesirable consequences of this transition by permitting additional early retirement. Hence, a decline in labor force participation (LFP) for older workers and a surge in the number of benefit recipients. Second, due to rising unemployment and declining LFP rates, the number of contributors to the pension system has eroded. Third, the pension system was not designed to deal with a society comprising a large share of private firms and a growing proportion of non-wage earners. Indeed, for tax administration purposes, the emergence of a private sector creates new challenges. In newly-created firms, employers and employees may try to a greater extent than previously to minimize their tax burden by taking advantage of existing loopholes. Indeed, fringe benefits (currently exempt from social security contributions) are becoming increasingly popular, shrinking the tax base for social security contributions.

Overall, assuming that employment rises, the ratio of contributors to pensioners could be stabilized by equalizing retirement ages for men and women and increasing both to 65 years. Moreover, the pension system should not bear certain costs related to redistributive/insurance functions (e.g., related to disability). The adoption of these measures could bring the pension system into balance. 1/ A new sustainable system would set benefit and contributions at levels compatible with a long-run, intertemporally balanced budget for Pension Funds. Such a long run balance would also promote the authorities’ broader fiscal objectives. 2/

A redefinition of the pension system should also seek to alleviate microeconomic inefficiencies which represent, in the long run, a serious threat to the stability of the pension system. These microeconomic inefficiencies are of two distinct types. First, the current pension system is exposed to unnecessary high costs due to arbitrage possibilities. That is, certain groups have an economic incentive to take undue advantage of certain provisions. For instance, contribution years can be purchased at a low cost by pensioners to increase their benefits. Second, the pension system may create distortions, altering behavior and, particularly, savings and labor supply.

Since the distinction between these two types of inefficiencies is somewhat blurred, an example may be useful. Various studies have suggested that seniority is becoming a more influential determinant of wages in Eastern European countries, possibly in an effort to cement long-term relations between employers and employees. 1/ Whatever its causes, the trend towards greater seniority makes pension benefits more costly because benefits are based upon the ten best yearly wages. Indeed, if employers and employees agree about a wage structure that provides relatively higher wages late in the life-cycle, benefits will be relatively higher. This has an arbitrage dimension. Such an arrangement also creates distortions because the life-cycle pattern of labor supply is altered and the higher benefits are financed by taxes levied on other contributors.

Two different approaches can be envisaged to deal with these various inefficiencies. First, a progressive phasing out of the PAYG system and a parallel reduction in social security contributions could be effected. A corollary, recently advocated in World Bank (1994), is a switch to a fully-funded pension system. An alternative is to minimize inefficiencies in the current PAYG system. Benefits would need to be tightly linked to contributions and provide a fair, market-related return to pensioners on previously-paid contributions. In this case, social security contributions amount to forced savings. Barring any liquidity constraint, if the return on these forced savings is high enough—that is if pension benefits are generous enough—social security contributions create nominal labor supply or savings distortions. 2/

In the short run, there are three options for tightening the link between contributions and benefits. First, the “ten best year” rule could be replaced by a longer period, say a phased move to thirty years. Second, early retirement could be progressively discouraged. Third, with a widening in the income distribution, a larger dispersion of pension benefits would be needed. Indeed, if pension benefits remain capped, marginal contributions from high-income earners are effectively taxed creating distortions. These reforms would be helpful even if, at a later stage, the PAYG system were to have a smaller role.

In the medium term, two principles could provide guidance to find the appropriate balance between a PAYG and a fully-funded system:

Rate of return. A PAYG system may be unable to deliver an appropriate level of benefits. Indeed, adverse demography or low productivity growth can lead to fairly low implicit rates of returns. Concerning Slovenia, once the micro-foundations—i. e., the design of pension benefits—are established, the return on a balanced PAYG system could be reasonably high, given expected productivity gains.

Risks. Even if the rate of return on pension contributions is low compared to the return on market investments, a PAYG system provides a hedge against certain macroeconomic or financial risks. This “portfolio diversification” argument is particularly powerful in transition economies. Indeed, these economies represent extremely risky financial environments for two reasons. First, volatility in financial markets is high. Second, the lack of a regulatory environment and well-established financial markets creates considerable risks for financial investors. Against this background, some simple simulations suggest that preserving a sizeable role for PAYG pensions could be welfare-enhancing.

I. The pension system in Slovenia: basic provisions and financing

This section provides a brief account of the administrative organization and of basic provisions of the benefit system (pensions and unemployment) in Slovenia. 1/

Pension insurance has a long history in Slovenia, dating back to the nineteenth century. Formally, the current scheme was established in a March 1992 law, yet it bears strong similarities to the pension system of the former Yugoslavia.

Administration. The pension system is administered by an autonomous body, the Institute for Pension and Disability Insurance of Slovenia (known under its Slovenian acronym ZPIZ). The ZPIZ has a staff of 680 employees in nine regional offices. Its managing body consists of representatives of employers, employees and pensioners. The State has the ability to determine the ZPIZ’s obligations, i.e., the disabled and pensioners’ rights. However, the ZPIZ is not under the formal auspices of the Ministry of Finance and independently sets contribution rates. So far this segmentation of responsibilities has not created conflicts, due to the cooperative attitudes of the main players.

Eligibility. Two conditions must be met before’a worker is eligible to receive a retirement pension. First, one must have contributed for 40 (35) qualifying years. 2/ Second, since 1992 a minimum pensionable age is required, initially set at 55 (50), progressively increasing up to 58 (53) by 1998. If the first condition is not met, one becomes eligible at 63 (58) with at least 20 qualifying years or at 65 (60) with 15 qualifying years. 1/ Early retirement benefits are available for disabled people and for the elderly and long-term unemployed. 2/ Benefits also cover those whose labor contract has been terminated due to the bankruptcy of their employers and are unable to find a job. Finally, pensions are also granted to survivors under similar eligibility conditions.

Pension Benefits, Once eligible for a pension, a worker’s benefit level depends on past earnings and accumulated qualifying years. To ensure comparability, past wages are indexed to the average wage growth in the economy. The pension base depends on average labor earnings during the best ten consecutive years. Pension benefits equal 85 percent of the pension base, multiplied by a correcting factor (0.866 imposed in 1991 to reduce all pensions); they are reduced by 2 percent for each year below the pensionable age. Thus a pensioner with 30 qualifying years would receive 0.866(85 percent - 10 × 2 percent) - 56.29 percent of his pension base. For early retirees, 1 percent per missing year is deducted from the pension base until the pensionable age is reached. Such a modest penalty induces early retirement, and only 32 percent of pensioners are entitled to full pensions (see Kuhelj, op. cit.).

The law determines a minimum and a maximum pension base. Equivalently, maximum and minimum old-age pensions with full entitlement represent 0.54 and 2.64 times the average net wage. In May 1995, beneficiaries of the minimum pension represented 27 percent of the total (see Kuhelj, op. cit.). Regardless of past contributions, an old age pension has a floor set at 22 percent of the average wage and old age pensioners can receive a means-tested supplement. As of March 1994, 12 percent of the pensioners received this social supplement equalling, on average, 10.2 percent of the average net wage.

Contributions. Pensions and health expenditures have been financed mainly by payroll taxes. 3/ Slovenia’s social security contributions are currently set at 44.7 per cent of the net wage, with 31 percent earmarked for pensions. This contribution rate, shared equally between employers and employees, has doubled between 1986 and 1993. As Table 64 indicates, such high rates are by no means unique in Eastern Europe, bearing witness to a well-developed social safety net during the communist era.

Table 64.

Social Security Contributions in Eastern Europe in 1994 1/

article image
Source: Business Eastern Europe (1994).

Employers and employees. In addition to pensions, Social security contributions finance health insurance as well as other benefits.

Exceptions. The pension system allows for numerous exceptions for occupational groups including farmers, self-employed and employers (see Stanovnik et al., op. cit.). Farmers have only been integrated into the general pension scheme since 1983. They can choose their own contribution base with a floor set at the minimum wage. The State complements the farmers’ contributions by an equal amount to compensate for the lack of employers’ contributions. Farmers, therefore, have no economic incentive to declare more than the minimum amount for periods not included in the best ten years. Given the likely understatement of their contributions, farmers are the main beneficiaries of the 22 percent floor. Self-employed and employers can also determine their contribution rate and their tax base. Stanovnik et al. (op. cit.) also point out that in the private sector, in-kind benefits and other perks are tax-exempt. According to the Tax Administration, more than 8 percent of the employees are now taking advantage of this provision. Finally, contribution years can be purchased for a fixed amount for years spent in the military or in higher education. These rules are clearly advantageous to high-income groups.

Inequality and Poverty. The pension system has been reasonably successful in containing poverty, thus distinguishing Slovenia from other transitional economies. Indeed, as of March 1994, less than 18 percent of old-age pensioners earned less than half the minimum wage.

II. Demography and labor market imbalances: relative impact and potential solutions

An ageing population will impose a growing burden on future generations. Yet, extrapolating from recent current trends overestimates this burden because the transition to a market economy is an atypical period. Indeed, it has profoundly affected participation rates by encouraging early retirement. At the same time, high unemployment reduces contributors to the pension scheme, straining the PAYG system. This section tries to disentangle these effects and simulates the evolution of the old-age dependency ratio under alternative assumptions. A robust conclusion is that an increase in the average retirement age of five to ten years would suffice to stabilize the financial condition of the pension plan in the face of an ongoing deterioration of demographic patterns.

1. Demography vs. activity rates

What caused the deterioration of pension finances between 1990 and 1994? As Table 65 indicates, this deterioration was primarily driven by the increasing number of recipients, rather than by more generous benefits. Indeed, real old age pensions decreased in line with real wages between 1991 and 1994.

Table 65.

Pensions—Number of Recipients and Purchasing Power

article image
Source: National Pension Administration (1993) quoted in Stanovnik et al. Kukar (op. cit.) and staff calculations.

Demography alone is an unlikely explanation for this retirement boom. Since the early 1980s, demographic indicators have remained stable, with the proportion of elderly (above 65) hovering around 11 percent. The recent influx of retirees denotes a change in retirement behavior. Since 1991, and after a marked decline in the 1980s, the labor force participation (LFP) rate of elderly workers (55 and older) has dwindled (Table 66). These developments were well characterized by Stanovnik et al. (op. cit.): “Early retirement was a social policy aimed at reducing the crisis of the labor market”. In many transitional economies, elderly workers have been induced to exit the labor force to reduce unemployment. 1/

Table 66.

Activity Rates by Age Group in Slovenia 1/

article image
Source: Popovic (1994) and SORS (1994).

For each gender, the last row is based on the results of the 1993 Labor Force Survey which included 4000 persons. This relatively small number makes the evaluation of participation rate for small groups less reliable.

Confronted with rapidly increasing unemployment (from 4.6 percent in 1990 to a peak of 14.4 percent in 1993), Slovenia has encouraged those aged 65 and above and still active to retire by providing relatively generous benefits. Slovenia has thus facilitated labor shedding while sheltering the unemployed from large welfare losses (see Orazam et al., op. cit.). 2/ Active employment policy, combined with relatively generous unemployment benefits have been two key elements of the safety net. For old workers, although the replacement ratios for unemployment benefits and early retirement pensions are roughly equivalent, retirement is a more attractive option. Indeed, pension duration, has no limits. As such, early retirement can be viewed as a hidden form of unemployment. As a result, for the 50–64 age group, labor force participation rates are unusually low for a middle-income country and the unemployment rate is about half the unemployment rate for the entire population (see LFS (1994)).

2. Prospects for the future

The ageing of the Slovenian population stems from a rising life expectancy and low fertility. Between 1970 and the early 1990s, life expectancy at birth for men and women has increased from 65.0 and 72.3 years to 69.4 and 77.2 years, respectively. The fertility rate has collapsed in Slovenia, as in other transitional economies. Since the early 1980s, births have declined by approximately one-third.

Little is known about the interplay between demography and economic or political uncertainty. Yet, two factors suggest that low fertility could be a lasting feature of Slovenian demography. First, low natality has persisted since the beginning of the post-Tito era. Second, fertility has declined for all women and, therefore, does not result from a decision to postpone childbearing later in the life-cycle. Total fertility currently stands around 1.35, well below the population reproduction rate.

The past decade of low fertility will inevitably have repercussions on the pension system. Simulations provide some approximate measures (Chart 16). The simulations assume that fertility rates will improve somewhat in the future and progressively return to their pre-1980s levels. 1/ This favorable assumption does not prevent a marked ageing of the population after 2005. Dependency ratios, defined as the ratio of the elderly to the active population, are almost halved by 2040. The recovery only begins around 2045 when low-birth generations retire.

CHART 16
CHART 16

SLOVENIA: DEPENDENCY RATIOS

Citation: IMF Staff Country Reports 1996, 120; 10.5089/9781451835588.002.A001

Source: Data provided by Slovenian authorities.

Yet the outlook may not be as bleak as suggested by these simulations. Dependency ratios do not gauge adequately the strain on the pension system due to low labor force participation rates and lower unemployment. A modified dependency ratio takes account of these two factors. As argued elsewhere, this is a more appropriate measure than standard dependency ratios. 2/ These simulations posit that labor force participation rates for the 50–65 age group rise to the average EU level by 2000 and that between 1995 and 2010 unemployment is reduced by 3 percent.

Currently, elderly workers retire earlier than they did previously. The ratio of active workers to pensioners, the (modified) dependency ratio, stands at around 1.7 and the average retirement age is therefore close to 56 years. The modified dependency ratio would stabilize if workers were to prolong their active life. Such a behavioral change is likely as labor market conditions improve, or if Slovenia were to adopt more stringent early retirement regulations.

To sum up, if unemployment recedes and labor force participation rise, adverse demographic trends would be neutralized.

3. Stabilizing the cost of pensions and unemployment benefits

Simulations were undertaken to calculate the average retirement age which would stabilize the dependency ratio over the long run. Further increase in the retirement age is required to offset the adverse effect of low dependency ratios around 2040 on the finances of the Pension Fund. To gain some insight, one may simulate the pattern of the Pension Fund debt. It is assumed that the share of wages in GDP, payroll taxes and the replacement rates remain constant for the next century. Under these assumptions, the Pension Fund’s financial balance depends exclusively upon changes in the dependency ratio. For a given level of interest rates, the Pension Fund’s accumulated debt/assets can be calculated until the end of the twenty-first century. Conversely, the dependency ratio or the retirement age required to balance the Pension Fund’s debt can be determined. Notice also that the pension scheme is no longer a pure PAYG system since it can temporarily run surpluses or deficits. Table 67 reports these findings.

Table 67.

Retirement Age Required to Stabilize Pension Benefits

(As a percent of GDP)

article image

Two different scenarios are envisaged. In scenario A, labor force participation in Slovenia for the 50–64 age group reaches EU levels which amounts to an increase of 15 percent. In scenario B, it is assumed that retirement age for men and women is equalized, which increases the LFP rate for this age group by another 15 percent. Interestingly, in both cases, the average retirement age has to be raised above 60 years to balance the Pension Fund’s finances.

Table 67 also outlines the importance of the timing of the reform and of interest rates. It shows that delays in raising the retirement age would require even more drastic changes. Indeed, if pension reform were postponed by 10 years, the average retirement age would have to be raised by another 6 months to compensate for the accumulated deficits.

Lower interest rates also worsen the Pension Fund’s situation. Indeed, in these simulations the Pension Fund implicitly builds up reserves until 2040, to cope with low projected dependency ratios. Therefore, lower interest rates are detrimental because the return on assets accumulated between 1995 and 2040 is reduced. This effect is sizeable since, at their peak these assets represent 60 percent of GDP.

Alternatively, one may envisage to balance the Pension Fund’s finances by lowering the replacement ratio. This could be achieved, for instance, by imperfectly indexing pensions to productivity gains. The behavioral implications of such indexation will be discussed in the next section. A pure accounting exercise suggest that a reduction by 30 percent of the replacement ratio is required if the retirement age is left unchanged.

Overall, given current trends on pension expenditures, there is little doubt that the average retirement age needs to be raised rapidly by some five to seven years. It is noteworthy that more drastic reforms are envisaged or implemented in most European countries. Slovenia’s slightly more favorable situation is a legacy of its communist past with women’s high LFP rates.

Before discussing the microeconomic implications of this reform in the next section, it is worth comparing the situation of pensioners in Slovenia with their EU counterparts. The required age for full pension entitlement is sixty-five for men in many European countries (the Netherlands, Spain, Germany). The suggested reform would therefore put Slovenian pensioners on an equal footing. Interestingly, some countries are also planning to eliminate gender-related differentials for the retirement age (Austria, Germany). Some countries are planning to reduce the possibilities of early retirement. For instance, France is currently scaling down the early retirement programs which where widely used to facilitate the restructuring of certain industries. This is motivated, in particular, by the budgetary cost of these measures.

III. Reducing microeconomic inefficiencies

Pensions play a key role in the distribution of income and, therefore, almost inevitably create distortions. Balancing the Pension Fund budget is therefore only one element of pension reform; minimizing distortions is another important element.

1. Some basic calculation on a PAYG system

A balanced PAYG system imposes tight constraints upon the level of benefits available to pensioners. The same pensioners have, during their working life, contributed to the pension system and these contributions may be regarded as “forced savings”. An important issue is the rate of the return on this forced savings compared to the return on financial investments.

Let B,W,L,P, and τ respectively denote pension benefits, gross wages, the number of employees and pensioners and the contribution rate. The basic accounting identity for a balanced PAYG system is, τWL = BP. Put differently, the replacement ratio is simply the product of the tax rate by the dependency ratio. As a simple example, if the working life averages 45 years and life expectancy at retirement equals 15 years in a demographically stable environment, the replacement ratio should equal three times the tax rate.

The rate of return on pension contributions is derived by calculating the discounted value of pension benefits and compared to the return on other savings. If they are equivalent, a PAYG pension system is said to be “actuarially neutral” or “fair”. Distortions generated by an actuarially fair PAYG system are minimal since individuals that are not liquidity constrained can accumulate negative savings, thereby offsetting the impact of pensions. As a first approximation, a PAYG pension system is actuarially neutral if the interest rate equals the sum of the growth rates in real wages and in the labor force, i.e., if r=v+n. If the interest rate exceeds v+n, the pension system is equivalent to a tax levied on wages, with an income effect because of the implicit transfer to current pensioners.

An important implication is that pension contributions only create distortions if the return on these pensions departs from actuarial neutrality. Formally, if τ is the pension contribution rate, the marginal tax rate affecting household labor supply decisions is τ(l-g(r-v-n)) where g(0)=1. Table 68 provides estimates of g as a function of the interest rate, assuming a working life of 45 years and a 15 year retirement period. This table reads as follows. If r exceeds v+τi by one percent, the effective tax rate is (l-g(l%))τ - 0.24τ.

Table 68.

A Measure of the Distortionary Impact of Pension

article image
Source: Perraudin and Pujol (op. cit).

It is, therefore, key to project productivity growth and real interest rates over the next several decades. Indeed, if one posits, as suggested in the previous section, that dependency ratios stabilize due to a postponement of retirement, n is close to zero. Medium-term scenarios for Slovenia envisage high productivity growth to catch up with Spain and Ireland (see IMAD (1995)). Over the same horizon, one may expect the real interest rate to converge at least partly towards German levels, as the risk premium attached to Slovenia dwindles. Overall, this suggests that productivity could equal or even exceed real interest rates.

Various caveats, however, applies. One, forecasts on productivity might be biased by recent productivity gains and it remains to be seen whether they will prove sustainable in the long term. The IMAD itself envisages a less optimistic scenario. Second, a large fraction of productivity growth may be due to higher capital accumulation and not to total factor productivity growth. In other words, the underlying assumption in the IMAD’s work may be that Slovenia currently suffers from capital under-accumulation. If this were the case, as discussed in the next section, fostering saving would be key and a PAYG system could be detrimental.

Overall, given these uncertainties, in what follows, it is assumed that r = v = 2 percent a year. As shown, in the next section, going further requires a more careful discussion of the role of uncertainty.

2. Modelling benefits

Pure PAYG pension systems do not exist. Pensioners’ rights are usually defined by complex formulas and pension benefits may fluctuate. As recent history indicates, pension schemes usually achieve financial balance by modifying contributions as well as benefits. Individuals also experience widely different career paths. Thus, the microeconomic design of pensions has to specify how wage patterns and employment duration influence pension benefits. But, these provisions themselves may have a feedback effect upon labor supply and wages over the life-cycle.

A simple model of pension benefits is as follows. As explained above, Slovene pension benefits for person i at time t essentially depend on Wi, the average wage in Slovenia, the number of working years, Ti, and the ratio, Mi, of wages earned to average wages in the economy over the best ten consecutive years averaging period. Thus,

B t , i = a ( .85 - b M a x ( T R = T i , 0 ) ) M i W t ( 1 )

TR is the number of qualifying years required for full entitlement. For Slovenia, the current values for parameters a and b are 0.866 and 2 percent. Note that equation (1) ignores some provisions of Slovenian pension benefits like a floor and a ceiling.

The value of Mi then depends mainly on the seniority rate, i.e. the premium paid by firms for experience. The limited evidence available indicates that experience was less rewarded in former CPEs than in western economies. But, recent studies indicate a changing attitude (see Flanagan (1995) for the Czech Republic, and Rutkowski (op. cit.) for Poland). In Slovenia, Vodopivec and Hribar-Milic’s recent findings suggests that seniority strongly grows for the last ten years before retirement. As a result, wages for the best ten consecutive years would be inflated. This bias could be reduced by increasing the reference period from 10 to 20 or thirty years. The following table which performs this calculates it assuming a working life spanning over 45 years.

Such a simple exercise constitutes an argument to lengthen the reference period for the calculation of pension benefits. The “best-ten-years” rule leaves the level of benefits exposed to variation in the seniority rate. It creates also an incentive for firms to increase wages more steeply before retirement. Such a behavior may explain Vodopivec’s findings. As Table 69 shows, these risks would be limited if the reference period covered the best thirty years.

Table 69.

Ratio of Average Wages During the Life-cycle and the Reference Period

article image
Source: Perraudin and Pujol (op. cit).

How do the Slovene benefit levels (given by Equation 1) compare to what pensioners would receive in a fully-funded system? To answer this question, the discounted values of contributions should be compared with the discounted value of benefits. The replacement ratio, currently set at 85 percent, should be set so as to equalizes discounted contributions and benefits. Table 70 summarizes the result of these simulations.

Table 70.

Total Return on Pensions

article image

The table reads as follows. If the seniority rate equals 2 percent and a worker retires after 40 years of work with a remaining life expectancy of twenty years, its replacement ratio should equal 52 percent, instead of the actual 85 percent. To describe pensioners’ current situation, it is probably appropriate to assume that the seniority rate was close to zero during their working life. Therefore, benefits levels are approximately actuarially fair for those workers who retire after 45 years of work. They are far too generous when they retire earlier.

Other findings are as follows. One, in order to offset the effect of seniority, it would be appropriate to lower the replacement ratio and/or to increase the reference period to 30 years, which is in line with the reforms planned or implemented in other countries. Two, existing penalties for early retirement are too low. Indeed, the retirement age that maximizes the discounted difference between contributions and wages are reported in the last column of the table. Such a calculation, implicitly posits that the marginal utility of leisure is nil. The optimal retirement age would be lower if leisure had positive marginal utility. This calculation suggests that the penalty for early retirement would need to be raised to 5 percent in order to achieve actuarial neutrality. To refine these calculation, the surviving spouse’s benefits would have to be taken into account. The replacement ratio could be lower to 70-75 percent from 85 percent.

These measures would eliminate incentives to retire earlier and would, therefore, be consistent with achieving a balance budget for the Pension Fund. Moreover, because the return on provided on past contributions is fair, distortions created on labor supply would be minimized. 1/ High incomes also deserve further comments. As already said, because pension benefits are capped, the return on marginal contributions is nil for high-income earners. Thus, distortions or incentives to avoid taxation are maximal. A potential solution could be to cap also their contributions or to funnel them into fully-funded schemes with market-determined return. Given the small number of contributors concerned by this cap, revenue losses for the Pension Fund should not threaten its financial balance.

IV. Pension reform, risk and financial institutions

In the conventional neoclassical world, two different arguments favor funded pension schemes and PAYG schemes: the distortions they create on the labor market and their impact on capital accumulation. The previous section has shed some light on the former; this section is concerned mainly with the latter.

In a deterministic framework, the link between pension provisions and capital accumulation has been thoroughly analyzed (see Blanchard and Fischer (1989) for a presentation). A PAYG system is likely to reduce capital accumulation. Whether this is a desirable outcome or not, from a welfare standpoint, depends upon the level of capital accumulation. At the optimum, theory holds, the “golden rule” is that the return on productive capital should equal the rate of population growth. But, as Diamond (1965) showed, a competitive economy can perfectly reach a steady state characterized by capital over accumulation which is signaled by too low a level of interest rates. The economy is, therefore, dynamically inefficient.

Testing the possibility of dynamic inefficiency poses, however, difficult issues. In a nutshell, the key problem is to define appropriately the rate of return on capital and to transform Diamond’s model to take account of uncertainty. A growing body of literature has already brought interesting results. First, a modified version of the “golden rule” exists whereby the return on capital should exceed population growth (see Russel (1991)). Second, under technological uncertainty, conditions required for dynamic efficiency have been derived by Abel et al. (1989) and Zilcha (1990 and 1991). Third and most importantly, it has become clear that a “central planner’s” role, in a uncertain environment, is not only to set output at its optimal level (like in the deterministic case) but also to distribute risks optimally across generations.

Such an approach creates a natural link with another strand of literature on financial institutions and risk-sharing (see Allen and Gale (1994) and (1995)). The premises of this approach is that financial institutions and financial instruments play a key role in the process of spreading risks and should also be so analyzed. These authors stress that industrialized countries with a similar level of economic development have favored extremely different financial environment. As a result, their populations are exposed to very different risks as well as different returns on their savings.

This sub-section makes use of these ideas to analyze how a PAYG system can play a role in diversifying risks. It also outlines some implications for the future of pension system in Slovenia.

1. The optimal balance between PAYG and a fully-funded pension system some theoretical elements

This section builds upon models developed by Blanchet (1990), Blanchard and Weil (1992) and Gale (1990). The discussion can be framed into a stochastic extension of the Diamond-Samuelson overlapping generation model.

Assume that a representative agent of the generation born at time t maximizes:

U ( C 1 , t ) + β ¬ E t U ( C 2 , t + 1 ) = C 1 , t 1 - ϒ 1 - ϒ + β ¬ E t C 2 , t + 1 1 - ϒ 1 - ϒ ( 2 )

where ϒ is the constant elasticity of risk aversion, and β determines the individual’s preference for consumption. He is endowed with one unit of labor and works during the first period of his life to earn a gross wage wt. He pays taxes τWt earmarked for the financing of a PAYG scheme and, in addition, saves σtWt. During the second period of his life, he retires and consumes his two pension benefits. They amount respectively to τWt+1(1+n) and Rt+1 σtWt where n is the (deterministic) population growth and Rt+1 the return on savings. Formally,

C 1 , t = W t ( 1 - τ - σ t ) C 2 , t + 1 = ( 1 + n ) τ W t + 1 + σ t W t R t + 1 ( 3 )

A representative firm uses a Cobb-Douglas technology and output per worker is given by:

Y t = ( 1 + μ t ) K t α ( α < 1 ) ( 4 )

where μt is a stochastic variable representing productivity shocks assumed to be identically normally distributed. To simplify the model, the capital depreciation rate is set equal to 1.

We further assume that the stochastic wages and interest rates are determined by,

R t = α ( 1 + μ t - θ t ) K t α - 1 ( 5 ) W t = ( 1 + μ t + θ t ) ( 1 - α ) K t α

where θt is also assumed to be identically normally distributed. In other words, Equation 5 posits that in addition to pure productivity shocks, wages and the rate of return on capital are subject to other shocks. This is a convenient way to model fluctuations in labor and capital shares. One may think about these two variables from a business cycle perspective. A stylized fact is that labor and wages only respond slowly to productivity or demand shocks. For example, during expansionary spells, employment growth lags behind output growth. Therefore, the share of capital in total GDP increases. This suggests a negative correlation between θ and μt.

Solving the first order conditions for utility maximization leads to the Euler equation,

( 1 - τ - σ t ) - γ = β - γ E t [ R t + 1 1 - γ ( σ t + ( 1 + n ) τ N t + 1 R t + 1 W t ) - γ ] ( 6 )

The expression in RHS ratio compares the respective returns on a PAYG system and on financial savings.

Since the capital stock at time t+1 equals the level of savings in the previous period, discounted by the demographic growth rate,

K t + 1 = σ t W t ( 1 + n ) ( 7 )

Using this equality and replacing wages and rate of return by their values from Eq. (5), one gets

( 1 - τ - σ t ) = β σ t 1 - ( 1 - α ) ( 1 - 1 γ ) ( α + τ ( 1 - α ) ) α 1 - 1 γ ( W t 1 + n ) ( α - 1 ) ( 1 - 1 γ ) A ( θ t + 1 , μ t + 1 , τ ) ( 8 )

where

A ( θ t + 1 , μ t + 1 , τ ) = ( E t [ ( 1 + μ t + 1 - θ t + 1 ) 1 - γ ( 1 + 2 ( 1 - α ) τ θ t + 1 ( α + ( 1 - α ) τ ) ( 1 + μ t + 1 - θ t + 1 ) ) - γ ] ) - 1 γ ( 9 )

Note that A equals one when there is no uncertainty.

One can finally determine the saving rate of the young as well as expected consumption during the retirement period,

E t C 2 , t + 1 1 - γ = ( R t + 1 σ t W t + τ ( 1 + n ) W t + 1 ) 1 - γ ( 1 - γ ( γ - 1 ) 2 V ( α ( μ t + 1 - θ t + 1 ) + ( 1 - α ) τ ( μ t + 1 + θ t + 1 ) ) K t + 1 2 α ( R t + 1 σ t W t + τ ( 1 + n ) W t + 1 ) 2 ) ( 10 )

where V(.) denotes the variance.

Equation 10 illustrates the trade-off behind saving decisions. On the one hand, the return on a PAYG system may be lower; on the other hand, the provisions of a PAYG pension system may allow to diversify risks, in particular, if the correlation between productivity shocks and income distribution is strongly negative.

Let us now turn to the level of contributions τ which maximizes the expected utility of a representative agent. Blanchet (1990) analyzes the impact of demography and shows that an unfunded system is relatively more preferable when demographic growth is negative. This counter-intuitive result can be explained as follows. When demographic growth is negative, the return on the PAYG system is lower making a fully funded system relatively more attractive. This is, however, only a partial equilibrium approach. Indeed, if the total population is dwindling, a risk of capital over-accumulation may be lurking which a PAYG pension system may alleviate.

Turning now to the interplay between risk and pensions, the results are more intuitive: a PAYG system can be a useful means to share certain risks across generations. In our context, unexpectedly high or low wages may have two different causes, a productivity shock (μt) or a distributional shock (θt). In the first case, wages and the return on savings are affected in a similar manner and risk-sharing is not an option. But, if a distributional shock causes a wage raise, current wage earners share this boom with retirees, who would otherwise be negatively affected, through the PAYG provisions. This insurance aspect may justify a more important role for a PAYG system than in the deterministic case.

This analysis is a variation of the analysis by Blanchard and Weil (op. cit.) and Gale (1992) on government debt issues. Blanchard and Weil, for instance, tried to pin down the requisites for the government to play a Ponzi game. They showed that if there is no possibility of risk-sharing between generations (i.e., in their model, if the consumption of the young and the old generations are perfectly correlated) the government cannot play a Ponzi game, no matter what the average riskless interest rate is. Conversely, when this possibility does exist running a Ponzi scheme is dynamically efficient.

2. Policy implications

A recent World Bank report embarked upon an ambitious task: defining the appropriate balance between funded and unfunded pension systems which would reconcile economic efficiency while protecting the elderly (see World Bank, op. cit.). It concluded that the optimal system should rely upon three different “pillars”: “a publicly managed system with mandatory participation and the limited goal of reducing poverty among the old; a privately managed, mandatory savings system; and voluntary savings” (p. xiv). The limited role assigned to the first (PAYG) pillar is a result of its return, expected to be notably lower than that on the other pillars. As suggested by the above analysis, the return on savings should not be the only yardstick for comparing the different pension systems. The risk dimension is also key.

How can one gauge these risks? Industrialized countries provide a useful benchmark. As Table 71 indicates, over the last thirty years, the standard deviation of total GDP and real wages’ has typically stood around 1 ½ percent. Since long term interest rates fluctuate more, the following calculations set the standard deviation of the rate of return at 8 percent for industrialized countries. 1/

Risk aversion also plays a pivotal role. A combination of theoretical and empirical problems renders any assessment of its level perilous. In particular, a questionable feature of the utility function used here is that it links risk aversion and intertemporal substitution. In fact, one may perfectly find other utility functions which could exhibit low intertemporal elasticity of substitution together with high risk aversion (see a formal presentation in Weil (1993)). This could explain, in particular, the low impact of interest rates on savings combined with households’ prudence as investors. We shall simply assume here that 7 lies between 1 and 3.

Table 71.

Slovenia: Standard Deviation for Selected Variables in G-7 Countries and in Slovenia 1/

(In percent)

article image
Source: Fiorito and Kollintzas (1994) and Staff calculations.

All standard deviations for G-7 countries are calculated using detrended variables for the period 1960-1989; for Slovenia the reference period is 1987-1994 for real GDP and wages and 1991-1994 for interest rates.

Nominal long term interest rate deflated by GNP/GDP deflator for G-7 countries, short term interest rates for Slovenia.

Against this background, what level of uncertainty could be expected in a country like Slovenia? As outlined elsewhere, Slovenia has enjoyed a relatively smooth ride on the bumpy road to a market economy. Still, the transition has caused considerable changes in output and real wages over the last four years. Real interest rates have been relatively more stable, due to indexation mechanisms. Such a high variance reinforces the need for institutions providing some sort of insurance.

For the future, two polar scenarios can be envisaged. The first one paints a situation of low inflation achieved through a tight pegging to EU currencies, and accompanied by high growth and high productivity gains. These developments would permit Slovenia to catch-up rapidly with EU standards, and, ideally, would also alleviate conflicts concerning income distribution. Yet, even under these favorable assumptions, the variance of wages and interest rates is likely to be higher than in G-7 countries. Indeed, the smooth integration of a small open economy to international trade requires flexibility in the labor market. Therefore, real wages will have to adjust, as a buffer to real shocks, as in Japan in the sixties and the seventies. To gain some insight into the implications for pensions, it is assumed in the following simulations that all standard deviations are twice as high as in G-7 countries.

A second scenario assumes that distributional shocks are more important. This could be the case for instance if growth were unable to deliver enough job creations to reduce unemployment or if inflation were not to be tamed. This could cause large swings in real wages or real interest rate in line with what many transition economies have had to cope with. In that case in addition to the increase in variance embedded in the first scenario, we also doubled θt variance.

Table 72.

Optimal Proportion of Funded and Unfunded Pension Systems

article image
ϒ denotes the coefficient of risk aversion (see above).

These simulation results cannot pretend to provide an estimate of the optimal balance between funded and unfunded schemes. 1/ They do provide, however, some intuition about how this balance is altered by risk. In particular, they show that a complete shift towards a fully funded system is suboptimal and that something between 20 to 30 percent of retirement savings provided by a PAYG system is reasonable. In short, the often lauded Chilean approach may be too risky under some circumstances.

If financial risk diversification is key, particular attention should be paid to two aspects in implementating private pensions:

First, the availability of a large range of financial instruments. At present, the Slovenes can mainly invest domestically in short-term bills which limits the usefulness of funded pensions. Indexation clauses have also sheltered savings from inflation related risks. In the long run, however, long bonds and equities are needed to create the appropriate combination of instruments to balance risks and return. 1/ For the same reason, households should be allowed to invest abroad. As exchange rate instability wanes, investment abroad would certainly permit to reduce certain risks and to circumvent the lack of well-established domestic financial markets in the short run.

Second, pension reformers often tend to ignore or to downplay the importance of regulatory requirements for the existence of efficient pension schemes. Ironically, shortly after the publication of the World Bank report supporting funded pensions, the MMM debacle in Russia provided a limpid illustration of the risk incurred in a “non-SEC” environment. By the same token, the fate of Maxwell pensioners is even more worrying in a country rightfully proud of its outstanding tradition in finance. A particular aspect of these regulations is to assure that competitive practices will prevent pension fund managers from charging their customers undue fees. 2/

To go beyond these general conclusions would certainly require the questioning of some particular assumptions of the neoclassical model. In particular, our framework downplays the potential beneficial impact of sizeable financial markets. While analyzing the Chilean pension system, Holtzmann (1995) has couched in an endogenous-growth model the following argument. Fully-funded pensions foster an effective allocation of capital by increasing financial markets’ depth. Therefore, standard neoclassical growth models may overestimate the reduction in marginal productivity linked with capital growth. This argument is of particular relevance when the banking system is under-developed and constrains the financing of investment, which is the case in Slovenia.

Another issue of importance is the origin of savings. In a neoclassical environment, savings only represent postponed consumption. This is admittedly an over-simplification of saving behavior and a more realistic description would also include intergenerational motives or a pure “capitalist” behavior. As Blanchet (op. cit) noted, this could modify considerably the conclusion of this model. In particular, in line with Pasinetti’s argument, he showed that, if a “capitalist” class exists, the aggregate saving rate in the economy is determined by its behavior and size. Therefore, funded pensions would have little impact on capital accumulation in this context. Interestingly, this corresponds to Holtzman’s (op. cit.) findings for Chile: the emergence of funded pensions did not increase households saving rates and accumulation is probably determined more by retained earnings from small businesses.

Bibliography

  • Abel, A.B., N. G. Mankiv, L.H. Summers and R.J. Zeckhauser (1991), Assessing Dynamic Efficiency: Theory and Evidence, Review of Economic Studies, 56, pp. 120.

    • Search Google Scholar
    • Export Citation
  • Abraham, K. and M. Vodopivec (1993), Slovenia: A Study of Labor Market Transitions, mimeo., World Bank.

  • Allen, F. and D. Gale, Financial Innovation and Risk Sharing, MIT Press, Cambridge, Mass. 1994.

  • Allen, F. and D. Gale (1995), A Welfare Comparison of Intermediaries and Financial Markets in Germany and the U.S., European Economic Review, 39 pp. 179209.

    • Search Google Scholar
    • Export Citation
  • Bélanger, G. (1994) Eastern Europe - Factors Underlying the Weakening Performance of Tax Revenues, IMF Staff Working Paper No. 94 Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Blanchard, O-J. and S. Fischer, Lectures on Macroeconomics MIT press, Cambridge, Mass. (1989)

  • Blanchard, O-J. and P. Weil (1992), Dynamic Efficiency, the Riskless Rate and Debt Ponzi Game under Uncertainty, NBER Working Paper No 3992.

    • Search Google Scholar
    • Export Citation
  • Blanchet D. (1990), Retraites par Capitalisation et par Répartition selon le contexte demographique: quelques résultats comparatifs, Annales d’Economie et de Statistique, 18, pp. 6390.

    • Search Google Scholar
    • Export Citation
  • Bole, V. (1994), Sterilization in a Small Open Economy: The Case of Slovenia, Working Paper No. 1, Economics Institute of the Faculty of Law, Ljubljana, Slovenia.

    • Search Google Scholar
    • Export Citation
  • Bole, V. and M. Gaspari, The Yugoslav Path to High Inflation, in M. Bruno et. al. (eds.), Lessons of Economic Stabilization and Its Aftermath, MIT Press, Cambridge, Mass., 1991.

    • Search Google Scholar
    • Export Citation
  • Buehrer, T.S., Can Trade Losses Explain the Current Recession in Slovenia?. Ph. D. Thesis, Graduate School of Arts and Sciences, Harvard University, Cambridge, Mass., 1994.

    • Search Google Scholar
    • Export Citation
  • Business in Eastern Europe, Various Issues 1994.

  • Calvo, G.A. and C. Vegh (1994), Inflation Stabilization and Nominal Anchors, Contemporary Economic Policy Vol. XII, pp. 3545.

  • Calvo, G.A., L. Leiderman and C. Reinhart (1993), The Capital Inflows Problem, IMF Paper on Policy Analysis and Assessment, PPAA/93/10.

  • Calvo, G.A., R. Sahay and C. Vegh (1995), Capital Flows in Central and Eastern Europe: Evidence and Policy Options, IMF Working Paper No. 95/57, 1995.

    • Search Google Scholar
    • Export Citation
  • Fiorito, R., and T. Kollintzas (1994), Stylized Facts of Business Cycles in the G-7 from a Real Business Cycles Perspective, European Economic Review, 38, pp. 235269.

    • Search Google Scholar
    • Export Citation
  • Flanagan, R.J. (1995), Wage Structure in the Transition of the Czech Economy, IMF Staff Working Paper No. 95/36.

  • Gale, D. (1990), The Efficient Design of Public Debt, in Dornbusch R. and M. Draghi (eds.), Issues in Debt Management.

  • Hendry, D. F. (1986), Econometric Modelling with Cointegrated Variables: An Overview, Oxford Bulletin of Economics and Statistics, Vol. 48, pp. 20111.

    • Search Google Scholar
    • Export Citation
  • Holtzman, R. (1995), Pension Reform, Financial Market Development and Endogenous Growth: Preliminary Evidence from Chile, mimeo., IMF.

  • IMF (1993), Slovenia: A Tax Reform Strategy, Technical Assistance Report by Fiscal Affairs Department, IMF, Washington, DC.

  • IMF (1995), Slovenia: Proposals for Improving the Public Expenditure Management System, Fiscal Affairs Department, IMF, Washington, DC.

  • Kalcic, M. (1995), Disability Security within the Pension and Disability Insurance Scheme of the Republic of Slovenia, mimeo.

  • Khan, M. S. and C. M. Reinhart (1995), Capital Flows in the APEC Region. IMF Occasional Paper No. 122 Washington, DC.

  • Koltay, J. (1994), Unemployment and Employment Policy in Central and Eastern Europe: Similarities and Differences, Acto Oeconomica, Vol. 46 (3–4), pp. 293313.

    • Search Google Scholar
    • Export Citation
  • Kozar, A. (1993), A Model of Slovenia’s Balance of Payments, Reviews and Analyses Bank of Slovenia, September 1993, pp. 724.

  • Kuhej, J. (1995), The Establishing and Following of Pension Assessment Bases for Pension Benefits, mimeo.

  • Lahiri, A. (1991), Money and Inflation in Yugoslavia, IMF Staff Papers, Vol. 38, No. 4 pp. 751788.

  • Mencinger, J. (1989), Economic Reform and Unemployment, Privredna kretanja Jugoslavije, Vol. 19, pp. 2338.

  • Mencinger, J. (1993), How to Create a Currency—the Experience of Slovenia Weltwirtschaftliches Archiv, pp. 418431.

  • Orazem, P. and M. Vodopivec (1995), Winners and Loosers in Transition: Returns to Education, Experience and Gender in Slovenia, World Bank, Policy Research Working Paper, No. 1342

    • Search Google Scholar
    • Export Citation
  • Orazem, P., M. Vodopivec and R. Wu (1995), Workers Displacement during the Transition: Evidence from Slovenia, mimeo.

  • Perraudin, W. and T. Pujol (1994), Framework for the Analysis of Pension and Unemployment Benefit Reform in Poland, IMF Staff Papers, Vol. 41, No. 4 pp. 643674.

    • Search Google Scholar
    • Export Citation
  • Pleskovic, B. and J. D. Sachs, Political Independence and Economic Reform in Slovenia, Ch. 6 in 0. J. Blanchard, A. Froot and D. Sachs (eds.) The Transition in Eastern Europe, Vol. 1, NBER Project Report, University of Chicago Press, 1993.

    • Search Google Scholar
    • Export Citation
  • Popovic, B. (1994), New 1980 and 1990 Estimates of Activity Rates by Age and Sectoral Distributions for Some Countries of Eastern and Southern Europe, Bulletin of Labour Statistics, No. 2, ILO

    • Search Google Scholar
    • Export Citation
  • Potocnik, J., M. Senjur and F. Stiblar (eds.) (1995), Approaching Europe: Growth, Competitiveness and Integration, Institute of Macroeconomic Analysis and Development, Ljubljana, Slovenia.

    • Search Google Scholar
    • Export Citation
  • Russel, C.L. (1991), Optimal Growth under Uncertainty, Economic Letters, pp. 3135.

  • Rutkovski, J. (1994), Changes in Wage Structure and in Returns to Education during Economic Transition: The Case of Poland, mimeo, Princeton University.

    • Search Google Scholar
    • Export Citation
  • Stanovnik, T., and S. Kukar (1995), The Pension System in Slovenia: Past Developments and Future Prospect, International Social Security Review, Vol. 48–1, pp. 3544.

    • Search Google Scholar
    • Export Citation
  • Statistical Office of the Republic of Slovenia, Statistical Yearbook. Ljubljana, 1994.

  • Statistical Office of the Republic of Slovenia, Labor Force Survey Results, No. 607, 1994.

  • Sziraczki, G. and J. Windell (1992), Impact of Employment Restructuring on Disadvantaged Groups in Hungary and Bulgaria, International Labour Review, Vol. 131, No 4–5 pp. 471496.

    • Search Google Scholar
    • Export Citation
  • Vegh, C. (1992), Stopping High Inflation, IMF Staff Papers, Vol. 39, No. 3 pp. 626695.

  • Vegh, C. and A. Hoffmaister (1995), Disinflation and the Recession Now versus Recession Later Hypothesis: Evidence from Uruguay, mimeo., IMF.

    • Search Google Scholar
    • Export Citation
  • Vodopivec, M. and S. Hribar-Milic (1993), The Slovenian Labor Market in Transition: Issues and Lessons Learned, World Bank Working Papers, WPS 1162.

    • Search Google Scholar
    • Export Citation
  • Vodopivec, M. (1994), The Effects of Democratic Determination of Wages: Theory and Evidence from Self-Managed Firms, World Bank Working Paper No. 971.

    • Search Google Scholar
    • Export Citation
  • Vodopivec, M. (1995), Determination of Earnings in Yugoslav Firm: Can it be Squared with Labor Management, Economic Development and Cultural Change, Vol. 41, No. 3 pp. 623632.

    • Search Google Scholar
    • Export Citation
  • Weil, P. (1993), Savings and the Permanent Income Hypothesis, Review of Economic Studies, pp. 367383.

  • World Bank, Averting the Old Age Crisis: Policies to Protect the Old and Promote Growth, Oxford University Press, New York, 1994.

  • Zilcha, I. (1990), Dynamic Efficiency in Overlapping Generations Models with Stochastic Production, Journal of Economic Theory, 52, pp. 36479.

    • Search Google Scholar
    • Export Citation
  • Zilcha, I. (1991), Characterizing Efficiency in Stochastic Overlapping Generations Models, Journal of Economic Theory, 55, pp. 116.

  • Zizmond, E. (1994), Economic Developments in Slovenia, Eastern European Economics, November-December, pp. 7599.

  • Zorc-Rents, V. (1993), The Methodology of Monitoring Revenues and Expenditures from Tourism in Slovenia’s Balance of Payments, Reviews and Analyses, Bank of Slovenia, September, pp. 724.

    • Search Google Scholar
    • Export Citation
1/

These comparisons are made on the basis of data provided by the Agency for Payments, Supervision and Information on registered household income. The coverage is narrower than NIPA as the data relate to socially- and state-owned enterprises and private enterprises with more than 3 employees.

1/

For a more detailed description of the sectoral composition of value added, see Slovenia: Recent Economic Developments, 1993 (SM/93/105).

1/

These surveys are conducted every year in May and the series on unemployment corresponds to IL0 definitions.

2/

Various biases exist for both measures. Some registered unemployed are not actively looking for a job and are not taken into account in the IL0 definition. The Labor Force Survey, however, uses a relatively small sample (5,000 households) and may not be totally accurate.

1/

It has been argued that some of this reorientation may not have been altogether desirable and was a “distress” response to the loss of the markets in the former SFRY (Bole (1994)).

2/

For example, Pleskovic and Sachs (1993) argue that, markets in the other republics of the former SFRY accounted for almost 25 percent of the sales of Slovenian enterprises. In 1991, the loss of markets in the former SFRY caused a 38 percent decline in exports.

1/

Kozar (1993) presents a detailed econometric estimates of the determinants of various components of the balance of payments using quarterly data for the period 1987–1992.

1/

Econometric estimates of demand for consumer goods imports using quarterly data for the period 1987–1992 (Kozar (1993)) produce an elasticity of 0.87 with regard to real wages and other sources of household compensation (measured in deutsche marks). The study also shows that the imports of semi-finished products are, predictably, sensitive to exports and (negatively) to domestic production; the elasticities are 0.84 and -0.66 (lagged one quarter), respectively.

2/

Prior to 1992, processing is included in services.

1/

In 1994, the component “purchase of foreign cash and checks from nonresidents” of export of travel services totalled US$409 million. The currency composition was as follows: Italian lira (37 percent), German marks (32 percent), Austrian schilling (17 percent) and others (14 percent). Reduced form estimates of the sensitivity of tourism receipts (Zorc-Reins (1993)) reveal the following parameters and elasticities (in parentheses): number of tourist arrivals at the border (1.3), index of the real effective exchange rate (-0.98) and the US$/DM exchange rate (𢄒2.4).

1/

Data in the capital account may be subject to measurement errors. Large net errors and omissions are reported for 1991 and 1992. In addition, the classification of transactions between the current and capital accounts is difficult. In particular, it is possible that tourism receipts may have been underestimated and the inflow of currency and deposits by households overestimated.

2/

For a comparative description and analysis of capital flows for selected economies in transition (Bulgaria, Czech Republic, Slovak republic, Hungary, Poland and Romania), see Calvo, Sahay and Vegh (1995).

1/

Net short-term capital outflow on account of the household sector reached US$410 million in 1990.

2/

It is important to bear two facts in mind for purposes of the analysis below:(1) the recovery In Industrial production began around the middle of the year, and (11) after falling by about 25 percent over 1991–92, registered household purchases grew by over 20 percent in 1993 and generated the consumption boom, which spilled over into imports. The inflows of currency and deposits may have provided the monetary basis for this boom.

1/

For Slovenia, the interest rate on foreign currency time deposits and for Germany, interest rates on 3 month time deposits less than 1 million marks.

1/

Figures under the coefficients are t-values; (*) indicates signifigant at 90 percent level of confidence and (**) indicates significant at 95 percent level of confidence.

1/

Tourism receipts are not significant in any of the regressions and the index total industrial production performs somewhat better than the index of capital goods production.

2/

All the measures of the level of economic activity were statistically insignificant. These equations were estimated without the proxies for measurement error.

3/

The variable PKF accounts for almost 80 percent of the capital account over this sample period.

1/

Total medium- and long-term debt of the former Yugoslavia stood at US$15.1 billion at end-1991.

2/

Between 1991 and 1994, debt owed to the World Bank for which republican final beneficiaries could not be identified was allocated using the Fund quota key (see below).

3/

Since June 1991 by the order of the National Bank of Yugoslavia (NBY), no payments had been made on debt owed to Paris Club creditors. Subsequent to the achievement of monetary independence in October 1991, Slovenia has permitted principal payments on previously rescheduled debt to Paris Club creditors on debt that it considered to be unambiguously attributable to Slovenian final beneficiaries. Payments have been made in accordance with the original loan terms and reschedulings. Payments on other debt have not been made.

1/

Slovenia also undertook to take up 9.9 percent of the liabilities under the Trade and Deposit Facility representing obligations of Slovenian final beneficiaries.

2/

For a more detailed description of the trade and payments system, see Annex.

1/

Bole (1995) and Mencinger (1993), two BOS Board members, describe these stages and provide insights into the thinking of policy makers during these early and middle stages.

1/

Total sterilization costs during 1992–94 were 2.6 percent of GDP. In 1992 sterilization costs were 0.5 percent of GDP, in 1993 they rose slightly to 0.6 percent of GDP. In 1994 they increased sharply to 1.4 percent of GDP.

2/

During 1994 holdings of foreign exchange by the banking system rose by an average of US$300 million per quarter compared to US$100 million per quarter during 1993.

3/

According to the annual report profits fell from 2.5 percent of GDP in 1993 to under 0.1 percent in 1994.

4/

These BOS bills paid a nominal interest rate with a removable coupon—the warrant that entitled the bearer to a discount on the purchase of foreign exchange or tolar denominated BOS bills if inflation was above the target level (13 percent). The BOS was also attempting to influence expectations with this instrument.

5/

Sales of BOS bills with warrants from June-Sept 1994 totaled SIT 31 billion while temporary foreign exchange purchases were SIT 25 billion and net REPOs were SIT 13 billion.

1/

The controls required borrowers to deposit 40 percent of foreign exchange loans with maturity of less than 5-years (not used immediately to import) into unremunerated tolar bank account.

1/

For demand deposits, reserve requirements were lowered to 12 percent (from 12.5 percent); for 31 to 90 day deposits reserve requirements were increased to 6 percent (compared to 3 percent previously); for 91 to 180 days deposits requirements were lowered to 2 percent (compared to 3 percent formerly); for deposits with maturity between 181 days to 365 days requirements were 1 percent (compared to 3 percent previously); and reserve requirements were shifted to zero for deposits of over 1 year maturity (previously 0.5 percent).

2/

Since April 1995 the foreign exchange cover regulations were set at: 35 percent of monthly foreign exchange transactions; 100 percent of foreign exchange demand deposits (previously 90 percent); 75 percent for time deposits up to 3 months maturity; 35 percent for 3 month to 1 year time deposits; and 5 percent of foreign exchange time deposits over 1 year. Slightly different requirements exist for accounts held by nonresidents. The foreign exchange minimum can be met by either foreign bank accounts, holdings of foreign currency and checks in process; interbank foreign exchange claims; BOS foreign exchange bills (up to 120 day maturity), and low risk foreign securities (up to a maximum of 25 percent of the total cover requirement).

1/

Short-term loans do not carry the same requirements as Lombard loans (see below). The availability of short-term loans to a bank is set monthly by the BOS based upon the reserve money program and the level of excess foreign exchange assets of a bank over minimum cover requirements.

2/

REPO operations involved purchases by the BOS of its foreign exchange bills with mandatory repurchase by the banks after 7 or 10 days. Temporary foreign exchange purchases were of one month duration and renewable. In the last quarter of 1994, the BOS invited banks to convert outstanding temporary foreign exchange transactions into permanent purchases.

3/

For a short period in early 1994, the BOS auctioned REPOs based on offered real interest rates.

1/

Up until December 1994 for one bank and May 1995 for the other.

2/

Large fluctuations in the monthly inflation rate during 1992–93 are attributable to uneven administered price changes that often accentuated seasonal variations. These fluctuations moderated when the Government smoothed out its pattern of administered price changes.

1/

The sixth month lag is positive and significant in both directions. In the case of Ml influencing inflation, most of the earlier lags are positive but insignificant. In the opposite direction, most of the earlier monthly lags were negative and insignificant. In both cases, the F test of significance of all the 6 lagged variables on the dependent variable was insignificant.

1/

The exchange rate acts as a proxy for inflationary expectations and relative rates of return.

1/

For a period in early 1995, interbank market rates were below tolar deposit rates.

2/

These rates were lowered on only three occasions: second quarter 1993, first quarter 1994, and second quarter 1995.

3/

The agreement limited Interest on household demand deposits to 60 percent of inflation (30 percent for enterprises); savings deposits to the inflation rate; time deposits up to 3 months to Inflation plus 2 percent; time deposits less than one year maturity to inflation plus 8 percent; and time deposits over 1 year to inflation plus 10 percent. While for some categories these rates were above prevailing rates, the largest impact was on savings and short ten time deposits.

4/

For instance in the fourth quarter of 1994 time deposit rates (over 1 year maturity) averaged 35 percent and in June 1995 the nominal rate fell to 13 percent. Loans for short-term working capital fell from 42 percent in the last quarter of 1994 to 18 percent in June 1995.

1/

For instance the spread between deposit and lending rates during 1992–94 reported in IFS for the Czech Republic was 6 to 7 percent, for Hungary was 7 to 12 percent, for Poland 2 percent, and for the Slovak Republic 5 to 6 percent.

1/

It is levied on wages, pensions and other incomes, while social security contributions and financial savings are tax exempt.

2/

For instance, gifts are tax exempt if they amount to less than 5 percent of total income; corporate cars are valued, on a monthly basis, at ½ percent of their book value.

3/

See “Slovenia:A Tax Reform Strategy”, IMF (1993). For instance, a company can borrow money and use the proceeds to buy government bonds. This financial strategy reduces its tax burden since interest received is tax-exempt while interest paid is deducted from profits. Also, a manager who is the main shareholder of a company may transform his wages into tax-exempt capital gains.

1/

In 1992, negative operating results in loss-making firms reached 19 percent of GDP; operating surplus in profit-making firms only amounted to almost 4 percent of GDP. Subsequently, losses shrank, to 7 percent of GDP in 1994 and profits edged up to 5 percent of GDP.

1/

Most civil servants live in Ljubljana, where the cost of living is higher.

1/

This includes a reduction in social security contributions for exporters at a total cost of SIT 15 billion in 1995–96.

1/

In the short run, this also raises the issue of a proper management of budget expenditures. Various suggestions on this front can be found in IMF (1995). This subsection only addresses the long term aspect of the problem.

2/

Private practice is becoming increasingly popular amongst physicians, in particular, because it provides a more advantageous tax status.

1/

The APPNI replaced the SDK which performed most of these functions in Slovenia and other former Yugoslav republics.

2/

The Ministry of Finance is already responsible for tax collection from individuals.

1/

Only a summary income statement was available to the staff. Thus, analysis of the factors behind banks profit performance is difficult.

2/

Reserves (with the BOS) and other assets accounted for 15 percent of total assets in 1994, while foreign exchange assets amounted to 26 percent of total assets.

3/

The largest rehabilitation bank (see below) had assets to liabilities ratios of 158 percent for foreign exchange and 20 percent for tolars, respectively.

1/

Provisioning standards are as follows: 10 percent for category B (30 days overdue); 25 percent for category C (over 31 days to 180 days overdue); 50 percent for category D (from 181 to 365 days overdue); and 100 percent for category E (1 year overdue) are required.

2/

In 1993, three banks were issued operating licenses, two branches of foreign banks were approved, and 7 licenses were issued allowing existing banks to expand their scope of operations.

1/

Analysis of inflation and money demand during the Yugoslav era are available in Lahiri (1991) and Bole and Gaspari (1991). Description of economic events and circumstances in Slovenia are available in Pleskovic and Sachs (1993) and Zizmond(1994).

2/

Payments on time deposits consisted of previous month’s inflation plus a real interest payment (for instance on deposits over 1 year the real rate of return was ¾ to 1 percent per month during 1992 to mid-1995 or 9 to 12 percent per annum; demand deposits were partially indexed at about half the previous month’s inflation.

3/

Normally, money-based sterilization policies cause interest rates to rise (as they did in Slovenia), and depress aggregate demand causing a recession (Vegh (1992) and Calvo and Vegh (1994)). Although beyond the scope of this Appendix, Slovenia underwent an economic recovery during 1992–95. After falling by 9 percent during 1991, and 5.4 percent in 1992, real GDP rose moderately by 1.3 percent in 1993, and by 5.5 percent in 1994. The authorities specifically designed monetary policy to support an economic recovery.

1/

Descriptions of monetary policy and analysis of events by two of the BOS board members are provided in Bole (1994, 1995) and Mencinger (1994).

1/

About one-fourth of the weighted RPI goods are subject to administered pricing. Freely determined market price inflation is referred to as core inflation in this paper.

1/

Unlike in most countries, r is a published rate. The nominal rate is not known until after the fact as all interest payments are indexed either to inflation or exchange rate changes.

2/

Inflation was included in a variety of formulations, with and without the change in the exchange rate. In all case the coefficients on inflation were insignificant. This is not too surprising given the high degree of indexation of financial assets.

1/

The unreported results for broad money are similar with somewhat lower t-statistics; the coefficient on inflation is statistically insignificant.

1/

Proposed by some Slovene economists in 1993 (see Stanovnik (1995)).

2/

Since 1991, Slovenia has managed to balance its general government budget. In addition to GDP’s fast recovery, two other factors have considerably helped. First, Slovenia no longer bears the important cost resulting from transfers to former Yugoslavia (around 9 percent of its GDP according to some estimates (see Buehrer (1994))). Second, unlike many transition economies, effective tax collection has prevented a collapse of revenues (see Bélanger (1994) for a discussion).

1/

See Rutkowski (1994) for Poland or Orazem et al. (1995) for Slovenia.

2/

Assuming competitive behavior on the labor market.

1/

More extensive descriptions can be found in Stanovnik et al. (1995), Kalčič (1995) and Kuhelj (1995) for the pension system, and in Abraham et al. (1993) for unemployment benefits. This section draws heavily upon these papers.

2/

Corresponding figures for women are given in parentheses after figures for men.

1/

These limits will be valid in 1997 and are currently one year lower.

2/

Several levels of disability are recognized depending upon the type and severity. Each level creates different entitlements.

3/

The State budget also provides transfers, in particular to cover the costs associated with specific groups (e.g., farmers).

1/

See Jack (1995) for the Czech Republic, Szirâczki and Windell (1992) for Hungary and Perraudin and Pujol (1994) for Poland.

2/

Initially, the Yugoslav Labor Code of October 1989 authorized layoffs but firing costs were often considered prohibitive. These costs were pared back in February 1991 by (i) reducing the advance notice requirement period from 24 to 6 months; (ii) some cost-sharing of severance benefits associated with mass layoffs with the budget; and (iii) tax cuts and subsidized energy granted to loss-making firms to avoid redundancies. In 1993, employment subsidies totaled 3.5 per cent of GDP. (see Vodopivec et al. (1993) and Koltay (1994)). Concerning unemployment benefits, only workers with 9-month continuous employment history were deemed eligible. The minimum period of entitlement is 3 months and increases to two years for workers with cumulative employment exceeding twenty years.

1/

. In line with the current official projections.

2/
See Perraudin and Pujol (op. cit.). The modified dependency ratio is defined as
MDR=LaborForceParticipants-unemploymentRetiredPopulation+kUnemployment(1)

k, the ratio of the cost to the system of an unemployed person relative to that of a pensioner, is close to one half, k is low because only approximately one third of the unemployed receive unemployment benefits.

1/

One may make a case that outside the reference period, contributions and benefits are not linked, thereby creating distortions. Simulations in Perraudin and Pujol (op. cit.) suggests that, with a thirty-year reference period, these distortions remain of limited magnitude.

1/

Certain arguments, however, point in the opposite direction. One, the return on equities is typically more volatile than the return on bonds. Two, standard deviation measures underestimates the effect of extreme, non-normal events like a market crash. Put differently, an investor would have needed more than twenty years to recoup from the 1929 crash.

1/

An obvious limitation is that a two-period overlapping generation model constrains the replacement ratio to one. This certainly exaggerates the amount of savings required for retirement purposes. On the other hand, it is also important to bear in mind that the model is calibrated for total savings and not only for financial ones.

1/

An important question which goes far beyond the scope of this paper is whether or not financial instruments should preexist the creation of pension funds.

2/

As anecdotal evidence, the only existing pension funds foresees that its operational expenses could represent up to 15 percent of its annual income. This is incomparably higher than the fees charged by similar funds in the US. The Chilean experience suggests, however, that economies of scale provide a significant room for improvement.

1/

Situation at end-June 1995.

2/

The tolar is divided into 100 stotin.

  • Collapse
  • Expand
Slovenia: Recent Economic Developments
Author:
International Monetary Fund