Journal Issue


International Monetary Fund
Published Date:
January 2000
  • ShareShare
Show Summary Details

I. Does Bulgaria Suffer from a Competitiveness Problem? A Look at Real Exchange Rate Indicators and Export Performance1

A. Introduction and Summary

1. Bulgaria’s experience over the past couple of years has brought up the question of whether the country suffers from a competitiveness problem. In response to a severe financial and macroeconomic crisis, Bulgaria had adopted a currency board in mid-1997. After an initial stabilization-induced rebound in the second half of that year, activity and exports stagnated (Figure 1). Moreover, the current account position deteriorated, from a sizable surplus in 1997 a deficit of over 7 percent of GDP in the first half of 1999 (Tables A42 and A43). Only the second half of 1999 saw a resumption of growth and an improvement in the current account. For the full year 1999 GDP growth is estimated at 2.5 percent, with a current account deficit of 5.2 percent of GDP that was fully covered by foreign direct investment and other inflows. While a combination of adverse shocks, such as the Kosovo crisis, the restructuring of industries, and the global financial crises, certainly explain some of this mixed performance, one can also ask whether Bulgaria’s underlying competitiveness was adequate during this period.

Figure 1.Bulgaria: GDP, Exports, and the Current Account

Source: IMF staff calculations based on data from BNB and NSI.

2. This paper assesses key aspects of Bulgaria’s competitiveness. There are various ways of defining the “competitiveness” of an economy. The broadest definition relates to the sustainability of a country’s current account position. In that regard, Appendix V of the recently issued staff report (EBS/00/45) suggests that under the presently projected external environment and with continued sound economic policies Bulgaria’s medium-term external position is sustainable. This paper adopts a narrower focus, but one that is highly relevant for assessing Bulgaria’s recent experience and near-term prospects. Specifically, this paper examines the behavior of a variety of real exchange rate indicators as well as export performance. The main conclusions of this examination are summarized in the remainder of this introductory section.

3. The evidence from price and wage indicators does not suggest a misalignment at present. The observed real exchange rate appreciation since mid-1997 is low compared to the experience of other transition economies at Bulgaria’s stage of transition, and can to a significant extent be traced to the Balassa-Samuelson effect (faster productivity growth in the tradables sector than in the nontradables sector). Other factors that help explain the appreciation are price liberalization and increased levels of sustainable government expenditures and capital inflows. Regarding wages, measured in U.S. dollars they are among the lowest in the region and below the levels predicted by Bulgaria’s GDP per capita. Moreover, productivity growth has helped unit labor costs in U.S. dollars remain broadly unchanged over the past two years. Also, Bulgaria’s minimum wages appear to be at prudent levels: the ratio of minimum to average wage at one third is comparable to that in other transition countries.

4. Bulgaria’s modest export performance in the last decade, including the slump in 1998–99, appears to be mainly attributable to external shocks and a late start of sustained reform efforts. An initial spurt in exports induced by the opening of the economy from 1991 stalled by the middle of the decade, largely reflecting the absence of sustained stabilization and reform efforts. Once the economy was stabilized with the introduction of the currency board in 1997 and critical reforms finally began to be undertaken in 1998–99, Bulgaria was hit by a fall in partner country demand and export prices due to the global financial crises, and by trade disturbances related to the Kosovo crisis. These external shocks coincided with disruptions caused by the long-delayed start of intensive enterprise restructuring which had a major effect on the traditional export industries. Despite the shocks, Bulgaria managed to broadly maintain its market share in the EU during this period. Moreover, in the second half of 1999 exports started to pick up.

5. Taken together, this evidence indicates that Bulgaria’s competitiveness is currently not under major threat. The behavior of prices, wages, and exports suggests that the difficulties that Bulgaria has been experiencing were to a large extent attributable to a series of internal and external shocks, and did not reflect a fundamental competitiveness problem. This sanguine view regarding competitiveness is supported by the ongoing economic recovery and an improvement in the current account position now that the effects of the adverse shocks are tapering off.

6. If competitiveness is currently adequate, how can it be preserved and even enhanced? Safeguarding competitiveness is particularly important for a country with a currency board, like Bulgaria. With active monetary and exchange rate policy ruled out, there are fewer policy instruments and less margin for error than under other exchange rate regimes. Hence, it is vital for Bulgaria to persevere with the sound policies pursued over the last two years. Fiscal policy needs to remain prudent, with sufficient flexibility to react to any unforeseen developments. In the absence of hard budget constraints in the public sector, Bulgaria needs to continue with its strict incomes policy for state-owned enterprises. Labor market flexibility needs to be enhanced. The momentum concerning the restructuring of public sector enterprises needs to be maintained. To attract foreign direct investment and increase productivity, the quality of public services should be improved, red tape reduced, and existing laws enforced more stringently.

B. What Do Real Exchange Rate Indicators Tell Us about Bulgaria’s Competitiveness?

Price-based real exchange rate indicators

Basic concepts

7. Two basic concepts of the real exchange rate (RER) can be found in the literature. The external real exchange rate looks at the nominal exchange rate adjusted for differences in price levels between countries.2 This concept is rooted in the purchasing power parity (PPP) theory. The external RER is also the competitiveness indicator of the well-known Mundell-Fleming open economy model, where each country produces a single good which is in imperfect competition with the goods produced by other countries. The internal RER focuses on the relative price of nontradables to tradables within a country as a measure of the incentives for consuming or producing tradable as opposed to nontradable goods. This latter RER definition is frequently used in the context of small developing countries whose terms of trade can be considered as given.3 The two measures of the RER will often, but not always, move in the same direction.

8. There are also various notions of the equilibrium RER. When discussing the external RER rate, two equilibrium conditions are conceivable. The first is either relative or absolute PPP. The second is given by the RER that ensures a sustainable current account path. Related to the latter concept, the internal RER is usually discussed within a medium-term framework in which the external asset position of a country is held constant (see Montiel, 1999). The equilibrium RER is then the relative price of tradables to nontradables that insures full employment of resources with this external constraint. A disequilibrium can in principle take two forms. The RER may not be in line with its medium-term level because price adjustment is sluggish owing to some rigidities in the economy. Alternatively, macroeconomic policies or consumers’ behavior may be inconsistent with a viable external position. For example, consumption levels may be unsustainably high for maintaining external balance: even though prices are fully flexible, ensuring that the RER attains a short-run equilibrium, the RER can be misaligned with respect to its long-run equilibrium level.

9. The theory of the internal RER makes a number of precise predictions about the determinants of the equilibrium RER. Higher productivity growth in the tradables than in the nontradables sector will result in an equilibrium appreciation of the RER (this is the so-called Balassa-Samuelson effect; see Balassa, 1964, and Samuelson, 1964). A shift in spending propensities toward nontradables will similarly yield an equilibrium appreciation; an example is a higher level of (sustainable) government spending, since fiscal expenditures are typically biased toward nontradable goods. A worsening of the terms of trade will usually require a depreciation to maintain equilibrium. Other factors that will influence the equilibrium level of the RER are the world interest rate and the country’s openness to trade.4

10. While the aforementioned factors may move either way, in the case of transition economies an equilibrium real appreciation is likely. In these economies, the Balassa-Samuelson effect is typically substantial. Also, as pointed out by Halpern and Wyplosz (1997), there are several specific reasons to expect countries to experience equilibrium real appreciation during the transition. Increased demand for services is likely to result in an appreciated RER. Prices of utilities, previously set at unsustainably low levels, will tend to increase along the transition path and also warrant an appreciation. Higher public infrastructure spending and changes in the tax structure will similarly result in an increase in the relative price of nontradables. Larger levels of sustainable capital inflows associated with increased creditworthiness along the transition will result in a higher sustainable current account deficit, implying an appreciated RER.5

Empirical issues

11. This paper adopts an eclectic approach to assessing whether Bulgaria’s RER is misaligned. Empirically, the estimation of equilibrium RER has proven difficult for developing and industrial countries, and these problems are compounded for transition economies. Simple PPP comparisons, while often constituting an informative starting point, are of limited usefulness. First, they require knowledge about a historical date at which the RER was in equilibrium. Second, the approach does not take into account that the equilibrium (internal) RER may shift over time due to the aforementioned factors. More sophisticated methods usually require a considerable amount of inputs, which represents a major obstacle in the case of transition economies.6 For these countries, short consistent time series, limited sectoral data, and the lack of a clear benchmark equilibrium point are among the key difficulties.7 Therefore, while the paper makes use of the results from earlier crosscountry econometric studies, no new econometric analysis is attempted in this section. Instead, the section first presents historical developments in various measures of the internal and external RER, and compares the experience of Bulgaria to that of other economies at a similar stage of transition. Subsequently, an attempt to assess the significance of the Balassa-Samuelson effect analytically, and the relevance of a variety of other factors is discussed less formally.

Bulgaria’s experience

12. Bulgaria’s RER has been on a generally appreciating tread throughout the transition. At the beginning of the transition process, the (external) real effective exchange rate (REER), calculated as a weighted external exchange rate index using the CPI’s of trading partners and competitor countries, was on a steep appreciating path (upper panel of Figure 2). However, this appreciation proved unsustainable, and after a large widening of the current account deficit, the exchange rate collapsed in early 1994. Over the period June 1994-December 1995, the REER appreciated at a lower rate. The following year and the first half of 1997 were marked by increasing macroeconomic and monetary instability approaching hyperinflation, which resulted in a total loss of confidence in the currency and dramatic weakening of the nominal effective exchange rate and, consequently, the REER. The stabilization of the economy following the introduction of the currency board in mid-1997 brought about a strengthening of the nominal exchange rate, and the REER quickly rose to levels above the ones prevailing before the collapse. Since July 1997, the REER has appreciated further by 15 percent.8 A similar conclusion is obtained by looking at the development of the internal RER (lower panel of Figure 2). A proxy for the internal RER was obtained by computing the relative price of services (to a large extent nontradables) to industry (composed mostly of tradables), using sectoral deflators. While the series is not long enough to allow for an analysis of the whole transition period, the index started to appreciate in 1997, with the cumulative appreciation since mid-1997 amounting to approximately 20 percent.

Figure 2.Bulgaria: The Internal and External Real Exchange Rate

Source: IMF staff calculations based on NSI data.

13. The appreciation during the currency board period has been below the historical trend, suggesting no obvious misalignment. The rate of RER appreciation since mid-1997 has been slower than the initial trend in 1992–3, and also slower than the trend between June 1994 and end-1995. Of course, the relevant question remains whether the current trend is sustainable. In that regard, it may be useful to evaluate Bulgaria’s experience in the light of what has been observed in other transition countries.

14. The RER appreciation experienced by Bulgaria since mid-1997 seems quite moderate compared with other transition economies. This conclusion is based on a comparison with the appreciation seen by other transition countries over a period of three years, taking as a starting point the year in which they were at a similar stage in the transition process as Bulgaria in mid-1997. To identify the stages of transition of other economies that can be compared to that of Bulgaria in 1997, a comprehensive liberalization index developed by Denizer, De Melo, and Gelb (1996) is used.9 For Bulgaria, this index, which can range between zero and one, took the value of 0.67 in 1997. For 12 other transition economies, the year in which each country’s index was closest to 0.67 was identified and labeled as year “zero”. Figure 3 indicates that according to this comparison, Bulgaria’s appreciation since mid-1997 cannot be considered out of line; if anything, the pace of appreciation has been quite moderate.10

Figure 3.Bulgaria: REER (CPI-based) in Selected Transition Economies (Year Zero=100)

Source: IMF Staff calculations based on data from national statistical agencies.

Wage-based real exchange rate indicators

Basic concepts

15. A different set of indicators of competitiveness is based on wages rather than prices. Since they are directly comparable across countries, U.S. dollar wages in the manufacturing sector are widely used to assess competitiveness. Theoretically, there is a clear link between wages and the internal and external real exchange rates. For example, in a simple two-sector model, if the tradables sector is more capital intensive than the nontradables sector, there is a positive relationship between the internal RER and dollar wages.11 Similarly, higher wages are associated with higher price levels, yielding a positive relationship between wages and measures of the external RER. Moreover, wages corrected for differences in labor productivities—unit labor costs (ULC)—offer a good measure of profitability in the traded goods sector. By using ULC indices from various countries, one can compute measures that are similar to the external REER’s discussed above.

16. However, ULC measures also possess drawbacks. While they are useful for assessing a country’s ability to export, they are less informative regarding the broader determination of the current account. Since ULCs ignore the role of capital, they provide a misleading picture if productivity gains are the result of capital substitution for labor. Moreover, ULCs are quite sensitive to cyclical movements in labor productivity.12 Also, measures based on ULCs may provide incorrect information about profitability if there are intermediate goods in the production process. An additional empirical difficulty in the case of Bulgaria is the absence of comprehensive and accurate data on private sector wages.

Bulgaria’s experience

17. The wage-based indicators tell much the same story as the price-based indicators: Bulgaria’s present competitiveness does not appear to be out of line. In recent years U.S. dollar wages, ULCs, and the ULC-based REER have moved along the lines of the CPI-based REER, suggesting no obvious loss of competitiveness (Figure 4).13 Following an abrupt decline in 1996–97, U.S. dollar wages recovered after the stabilization of the macroeconomy and continued to increase steadily until March 1999. Since then, U.S. dollar wages have remained stagnant, in part reflecting a strong appreciation of the U.S. dollar against the euro (to which the lev is pegged). With productivity increasing, U.S. dollar unit labor costs declined by as much as over 20 percent during 1999, while the ULC-based REER has been broadly unchanged since late 1998.

Figure 4.Bulgaria: Wage Indicators.

Source: Staff calculations based on data from NSI.

18. Bulgaria’s wage levels continue to be low compared with other transition countries. Among the countries for which comparable data are available, only Albania, Romania, Russia, and Ukraine report lower levels (Table 1). Also, among southeast European countries Bulgaria’s budgetary sector wages are the third lowest after Albania and Romania (Table 2).

Table 1.U.S. Dollar Gross Industrial Wages in Selected Transition Economies
1998July/August 1999
Czech Republic367372
Slovak Republic293257
Source: IMF staff estimates based on data from national statistical agencies.
Source: IMF staff estimates based on data from national statistical agencies.
Table 2.Budget Sector U.S. Dollar Wages in Southeast European Countries


Source: IMF staff estimates based on data from national statistical agencies.
Source: IMF staff estimates based on data from national statistical agencies.

19. A cross-country comparison of wage growth also shows Bulgaria to be in a prudent range. As in the previous subsection, for each comparator country, “date zero” is defined as the year in which it had a similar score in the De Melo, Denizer, and Gelb (1997) liberalization index as Bulgaria had in 1997. U.S. dollar wage increases in Bulgaria since 1997 have remained below those experienced in Albania, Latvia, and Lithuania, but above those in Hungary, Romania, and Russia since their year zero. Of course, this comparison intentionally ignores the differing starting levels (wages in Hungary were nearly three times as high as Bulgaria’s in year zero, but those in Lithuania were only half as large) and should be interpreted cautiously.

Figure 5.US$ Wages and PPP-per Capita Income

Source: Staff calculations based on estimates by Krajnyák and Zettelmeyer (1998).

Figure 6.Wage Gap

(Estimated Equilibrium U.S. Dollar Wages as a Fraction of Estimated Equilibrium Wages)

Source: Staff calculations based on estimates by Krajnyák and Zettelmeyer (1998).

Figure 7.US$ Wages in Selected Transition Economies (Year 0=100)

Source: IMF staff estimates based on data from national statistical agencies.

Minimum wages

20. The present level of Bulgaria’s minimum wage does not appear to be a major threat to competitiveness. During the near-hyperinflation of 1996–97, the real value of the minimum wage had eroded substantially, reaching a low point at the equivalent of US$5 per month in early 1997 (Figure 8). Subsequently, the minimum wage was raised faster than other wages, and after a recent increase from US$35 to US$39 per month from February 2000, the ratio of minimum to average wage is now comparable to that in other transition economies (where it is typically one third, see Table 3). The government has decided that this ratio will be maintained in the future, through semiannual adjustments in the minimum wage in line with increases in public sector wages. The economywide significance of the minimum wage does not seem to be very large, since few workers (including only 1 percent of budgetary sector employees) actually receive it and given that the only remaining indexation is the link to minimum social security contributions.

Figure 8.Bulgaria: Real Minimum and Average Public Wages (1995=100)

Source: NSI.

Table 3.Minimum and Average Wages in Selected Transition Economies, 1999


Average Wage in US$117**372293*269425128114*25776*
Minimum Wage in US$3410473*86164299*9616*
Ratio Min-Average0.


Average Public Wages

Source: IMF staff estimates based on information from national statistical agencies.


Average Public Wages

Source: IMF staff estimates based on information from national statistical agencies.

Possible explanations for the observed RER appreciation

21. While the observed real appreciation in Bulgaria in recent years appears moderate, it is important to ascertain that it reflects the fundamentals and not a misalignment. To further probe the issue of whether Bulgaria’s competitiveness is adequate, this section takes a look at the behavior of some key fundamentals that may explain the observed RER behavior. While empirical problems preclude a quantitative assessment of the relative importance of these factors, the evidence indicates that the Balassa-Samuelson effect, price liberalization, and to some extent changes in capital inflows and government spending likely contributed to the RER appreciation since 1997. By contrast, a deterioration in the terms of trade and a reduction in import tariffs put some downward pressure on the RER in 1999, possibly preventing a further appreciation in that year.

The Balassa-Samuelson Effect

22. The Balassa-Samuelson effect refers to the impact of differential productivity growth rates in the tradables and nontradables sectors on the internal RER. Typically, productivity growth tends to be faster in the traded goods sector. If wages are equalized in the two sectors, this means that the relative price of nontraded goods will tend to increase. To the extent that the observed relative increased in the internal RER can be traced to such productivity differentials, it can be regarded as an equilibrium phenomenon, and does not imply a loss in competitiveness. Formally, in a simple model of a small open economy with Cobb-Douglas sectoral production functions, the relative price of nontradables can be expressed in the following way:

Here p, A, δ, and γ denote prices, total factor productivity, and the labor shares in the nontraded and traded sectors, respectively, and the subscripts T and NT stand for traded and nontraded sectors. This relationship is expected to hold in the long run; hence demand factors and temporary wage differentials due to imperfect labor mobility across sectors are ignored.

23. Even for industrial countries, this effect has been estimated to be significant.14 In the case of Portugal, Swagel (2000) estimates that the Balassa-Samuelson effect could account for as much as 2.5 percentage points of the annual inflation. For transition economies, the effect is likely to be larger.

24. What role has the Balassa-Samuelson effect played in Bulgaria? Given data problems, precise estimates are hard to come by, but the evidence points to this effect being a main factor responsible for Bulgaria’s real appreciation. This is demonstrated in Figure 9, which shows the relative price of nontradables as predicted by equation (1), together with the actual evolution of the price ratio.15 The actual and predicted price ratios tend to move in tandem. In fact, until the second half of 1999, the calculations indicate that the observed real appreciation could be fully explained by changes in relative productivity. In the third quarter, however, output in the service sector grew much stronger than in the industry sector, and this was not fully reflected in a corresponding fall in relative prices. As a result, the predicted RER stood at 6.5 percent lower than the actual RER. Keeping in mind, however, that labor productivities are bound to fluctuate in the short run, and that the model is intended to capture long-term trends, the calculations provide clear support for the view that a substantial fraction of the observed appreciation can be traced to shifts in productivities.

Figure 9.Bulgaria: The Relative Price of Nontradables to Tradables—Actual and Predicted

Source: Staff calculations based on data from NSI.

Other factors explaining appreciation

25. Other factors—whose importance is more difficult to assess quantitatively—are also likely to have contributed to the observed appreciation since mid-1997. First, to the extent that administered prices, particularly in the utilities sector, have been set below market prices, their liberalization results in an equilibrium increase of nontraded goods prices. In Bulgaria, prices were liberalized substantially at the beginning of 1999, when the share of administered prices in CPI was reduced to 14.3 percent. In the same year, these prices increased by 22 percent whereas the overall CPI increase was just 6.2 percent. Second, government expenditures (weighted toward nontradables) have increased markedly since the crisis of 1996/97 (Figure 10), which should have resulted in some equilibrium appreciation. Finally, Bulgaria’s improved capital account position has created room for an equilibrium appreciation: a net outflow of 9.4 percent of GDP in 1996 had turned into a net inflow of 6.2 percent of GDP in 1999.

Figure 10.Government Expenditure and the Terms of Trade

Factors working against appreciation

26. Some other factors are likely to have exerted downward pressure on the RER, particularly in 1999. Over the last three years, Bulgaria has reduced average tariff rates from 16.8 percent in 1997 to 13.7 percent from the beginning of 2000. Non-tariff barriers were also substantially reduced: import surcharges have been eliminated, all but one export ban lifted, export taxes removed, and import licensing abolished. The direct effect of these measures is to put pressure on the equilibrium RER to depreciate. However, trade liberalization also has indirect effects that can be substantial: a more open trade regime is likely to enhance the adoption of more advanced technologies, increasing productivity and enhancing the Balassa-Samuelson effect. Movements in the terms of trade represent another factor: while Bulgaria’s terms of trade improved in 1998, they worsened in 1999 (Figure 10). This exerted downward pressure on the equilibrium RER in 1998 and upward pressure in 1999 when the real appreciation indeed stalled.

C. Bulgaria’s Export Performance

27. This section looks at export performance, perhaps the most direct test of competitiveness. In Bulgaria’s case the assessment is complicated by adverse shocks whose effects are difficult to disentangle from those of a possible loss in competitiveness. The picture that emerges is that throughout the 1990s Bulgaria’s export growth was rather modest, and the country made only limited inroads into the EU market. The performance in 1998–99 was particularly disappointing as exports actually fell in both years. This slump appears, however, to have been mainly attributable to severe, largely temporary shocks, and exports started to recover in the second half of 1999. While this would indicate the absence of an imminent threat to competitiveness, there is clearly a need to persevere with macroeconomic stabilization and structural reforms to consolidate these gains.

28. Bulgaria’s exports were relatively stagnant during the 1990s, especially when compared with the experience of the more advanced transition economies. In 1998–99, Bulgaria’s exports were only some 20 percent higher than at the beginning of the transition in 1991–92 (Figure 11). While this performance compares reasonably well with other countries in the region, it is modest when compared with the substantially higher growth rates in the more advanced central European transition countries (Figure 12).

Figure 11.Bulgaria: Export Performance During the 1990s

Figure 12.Export Performance of Selected Countries

1/ Data for Poland and Hungary for 1993-98, for Slovak Republic for 1994-98.

29. The overall picture of sluggish export growth masks three distinct phases. The first phase was marked by strong export growth of some 20 per annum during 1992–95. This was followed by a period of stagnation in 1996–97 at the time of the foreign exchange and financial sector crisis. Thereafter exports slumped during 1998–99, experiencing two successive years of negative growth at an annual average rate of–10 percent.

30. The sharp pick-up in exports during the first phase largely reflected the opening up of Bulgaria’s economy following decades of relatively limited trade with countries outside the former Soviet Union. The normalization of Bulgaria’s economic relations with western Europe provided a strong initial impetus for export growth, and helped to offset the effects of a slowdown in industrial country trading partners during 1992–3 and of trade sanctions on Serbia and Montenegro which blocked Bulgaria’s export routes during 1993. During the first half of the decade, Bulgaria also succeeded in diversifying its export markets away from its traditional partners toward the EU (Figure 13 and Table A47). The share of exports to the EU grew 26 percent during 1993–95, mostly at the expense of Russia, traditionally Bulgaria’s largest export partner. Also, Bulgaria’s market share in the EU improved by 50 percent during this period (Figure 14).

Figure 13.Bulgaria: Direction of Trade (Percent of Total Exports)

Source: Bulgarian authorities

Figure 14.Market Share in EU Imports

1/ Index of market shares of cumulative exports over four preceding quarters.

31. The stalling of exports during 1996–97 was largely attributable to insufficient progress on structural reforms. Indeed, this underlying weakness was already evident from the limited product differentiation of exports during the phase of high export growth (Figure 15). In the absence of restructuring of key export-oriented state enterprises accounting for the bulk of exports, Bulgaria’s exports continued to be dominated by machinery, chemicals, and metals, which comprised more than half of total exports in 1997 (Table 4 and A46).16 While some reforms were started during this period, the tentative nature of these reforms implied that the fruits of liberalization would be reaped only over time. Small initial steps were taken toward effectively restructuring enterprises which had been shut off from generous credit pipelines after the 1996 banking crisis. The reversal of interventionist agricultural policies which had hurt agricultural exports got off to a tentative start. Trade liberalization began, albeit at a slow pace initially (Tables A59–A61)1718. The delays in introducing reforms (and the lack of macroeconomic stability until the currency board was put in place in mid-1997) reduced investor confidence, explaining Bulgaria’s limited success in attracting foreign direct investment when compared to other transition countries in the region (Figure 17). According to business surveys, other factors such as corruption, bureaucracy, and administrative inefficiencies also acted as deterrents to foreign investment, and help explain Bulgaria’s relatively late start in attracting foreign direct investment flows.19

Figure 15.Bulgaria: Product Diversification of Exports (1993-98)

(Percentage of total exports)

Source: Bulgarian authorities.

Table 4.Enterprise Reform Index and Overall Reform Index
Enterprise Reform Index 1/Overall Reform Index

(1997) 8/
Czech Republic3Bulgaria0.67
Lithuania2.7Slovak Republic0.77
Slovak Republic2.7Slovenia0.79
Bulgaria2.3Czech Republic0.82
Source: EBRD Transition Report, 1999 and Growth Experience in Transition Economies (SM/98/228).

In 1998. Scores of “1” and “4” imply limited progress and good progress respectively.

Using an overall structural reform index, as constructed by de Melo et al. (1996) and updated using the EBRD structural reform index.

Source: EBRD Transition Report, 1999 and Growth Experience in Transition Economies (SM/98/228).

In 1998. Scores of “1” and “4” imply limited progress and good progress respectively.

Using an overall structural reform index, as constructed by de Melo et al. (1996) and updated using the EBRD structural reform index.

Table 5.Cross Country Comparison of Trade Parameters
Trade Restriction Index Score 1/Average tariff 2/Non-Tariff Barriers 3/
Czech Republic1116.91
Slovak Republic21371
Source: IMF.

As defined in EBS/97/163, Trade Liberalization in Fund-Supported Programs. The index consists of a 10-point scale that combines measurments of the restrictiveness of tariffs and non-tariff barriers. It measures the overall restrictiveness of a country’s trade system relative to protection levels in all Fund members. A rating of 1 represents the most open trade regime and 10 the most restrictive.

Source: IMF.

As defined in EBS/97/163, Trade Liberalization in Fund-Supported Programs. The index consists of a 10-point scale that combines measurments of the restrictiveness of tariffs and non-tariff barriers. It measures the overall restrictiveness of a country’s trade system relative to protection levels in all Fund members. A rating of 1 represents the most open trade regime and 10 the most restrictive.

Figure 16.Correlation Between Export Growth and Trade Reform Since the Beginning of Stabilization

Source: EBRD Transition Report 1999 and IMF staff estimates.

1/ The degree of trade liberalization is measured by the change in a country’s current score according to the IMF’s trade restrictiveness index, and the most “restrictive” score of 10. This assumes that the countries in the sample had very restrictive trade regimes at the beginning of stabilization. This assumption is justified by the fact that before stabilization began these countries had centrally planned trade and barter trade agreements.

Figure 17.Cumulative FDI Inflows (1989-98)

32. The slump in exports during 1998–99 appears to have been primarily the result of adverse external shocks and disruptions caused by the delayed start of intensive enterprise restructuring (Figure 1819). Two external shocks dealt a major blow to Bulgaria’s exports in 1998–99. First, the global financial crises weakened Bulgaria’s external environment markedly. Bulgaria’s export prices fell by 9 percent in 1998, and remained at that lower level in the following year (Table A44). At the same time, partner country demand was weak, contributing to negative export volume growth in both years. Second, in March-June 1999 the Kosovo crisis blocked transport routes to Bulgaria’s main export markets in Europe, causing delays, cancellations, and costly rerouting. The land routes were restored during the summer, but transit by ship along the Danube river remains blocked to date. In all, export losses to Bulgaria from the Kosovo crisis during 1999 are estimated at some $100 million, or 2½ percent of annual exports. However, the poor export performance in 1998–99 cannot be fully explained by the negative external shocks as other countries in the region, similarly affected by the shocks, out-performed Bulgaria during this period (Figure 20). An additional explanation lies in the fact that Bulgaria started a serious restructuring of its traditional export-oriented state enterprises only during 1998–99. This restructuring caused inevitable disruptions which exacerbated the adverse effects of the external shocks on exports. Reflecting the combined effect of all these shocks, the growth in Bulgaria’s market share in the EU faltered during 1998–99, and remained on a plateau from the second half of 1998.

Figure 18.Bulgaria: Export Price and Demand

(Annual Growth Rate)

Source: IMF World Economic Outlook and staff estimates.

Figure 19.Bulgaria: Monthly Exports

Figure 20.Exports by Selected Countries Since Mid-1998

(U.S. dollar million)

Source: IMF staff estimates.

33. A recovery in exports from mid-1999 suggests that the slump was not attributable to inadequate competitiveness. The recovery which has so far been rather gradual reflects the tapering off of the external shocks. While the disruptions from restructuring continue to some extent, some enterprises in the traditional sector—such as iron and steel—have resumed production for exports after being restructured or privatized (Figure 21). Indeed, for the first time since 1991, a discernible shift in the commodity structure of exports has become evident. While traditional commodities like metals, chemicals, and machinery still dominate, a pick up in the growth in some categories of non-traditional exports such as textiles has begun (Figure 22). There have also been recent signs of an improvement in Bulgaria’s market share in EU imports (Figure 23).

Figure 21.Bulgaria: Export Recovery During 1999

Figure 22.Bulgaria: Traditional and Non-traditional Exports

(In millions of U.S. dollars)

Source: Bulgarian authorities.

Figure 23.Bulgaria: Market Share in EU Imports (1998–99)

D. Implications for Economic Policy

34. It appears that Bulgaria’s competitiveness is not immediately under pressure despite the weak export performance over the past few years. Bulgaria’s success in preserving competitiveness in the face of external shocks can be attributed to the appropriate and timely policy responses taken by the authorities, and the acceleration of much-needed reforms in the last two years which have already begun to be yield dividends.

35. The challenge now is to stay on course and persist with policies that will maintain and even strengthen competitiveness. Bulgaria’s currency board arrangement allows it very few degrees of freedom, and therefore policy efforts need to continue on a wide front. The list of is would include, inter alia, completing the restructuring of export-oriented state enterprises, land reform, as well as structural reforms to improve the competitiveness of the economy, improve the business climate and support the operation of the private sector. The key priorities include the following.

  • to ensure that real wage increases are in line with productivity growth, including through continuing a strict incomes policy for state enterprises;
  • to remove labor market rigidities through amendments in the labor laws;
  • to complete the restructuring of the remaining large state enterprises and utilities;
  • to improve administrative efficiency through continued improvements in public administration and training;
  • to reduce red tape and corruption to attract foreign direct investment and support the private sector;
  • to improve financial sector intermediation; and
  • to streamline and enhance the existing legal and regulatory framework.

    Alberola-IlaEnrique and TimoTyrväinen1998“Is There Scope for Inflation Differentials in EMU?,”Documento de Trabajo No. 9823 Banco de España

    • Search Google Scholar
    • Export Citation

    BaffesJohnIbrahim A.Elbadawi and Stephen A.O’Connell1999“Single-Equation Estimation of the Equilibrium Real Exchange Rate” in: Lawrence E.Hinkle and Peter J.Montiel (eds.) “Exchange Rate Misalignment – Concepts and Measurement for Developing Countries,”Oxford University Press

    • Search Google Scholar
    • Export Citation

    BalassaBela1964“The Purchasing Power Doctrine: A Reappraisal,”Journal of Political Economy 72

    De MeloMarthaCevdetDenizer and AlanGelb1996“From Plan to Market: Patterns of Transition,”World Bank Policy Research Paper 1564

    • Search Google Scholar
    • Export Citation

    De GregorioJosé and HolgerWolf1994“International Evidence on Tradables and Nontradables Inflation”IMF Working Paper 94/33

    De GregorioJoséAlbertoGiovannini and ThomasKrueger1994“Behavior of Nontradable- Goods Prices in Europe: Evidence and Interpretation,”Review of International EconomicsVol. 2No. 3 pp. 284-305

    • Search Google Scholar
    • Export Citation

    DevarajanShantarajan1999“Estimates of Real Exchange Rate Misalignment with a Simple General-Equilibrium Model,” in: Lawrence E.Hinkle and Peter J.Montiel (eds.) “Exchange Rate Misalignment – Concepts and Measurement for Developing Countries,”Oxford University Press

    • Search Google Scholar
    • Export Citation

    European Bank for Reconstruction and Development (EBRD)1997–99“Transition Report,”London

    EdwardsSebastián1989“Real Exchange Rates, Devaluation, and Adjustment,”Cambridge, MA, MIT Press

    FischerStanleyRatnaSahay and Carlos A.Végh1996“Stabilization and Growth in Transition Economies: The Early Experience,”Journal of Economic PerspectivesVol. 10No. 2 pp. 45-66.

    • Search Google Scholar
    • Export Citation

    GaribaldiPietorNadaMoraRatnaSahay and JerominZettelmeyer1999“What Moves Capital to Transition Economies?,”unpublishedInternational Monetary Fund

    • Search Google Scholar
    • Export Citation

    HalpernLásló and CharlesWyplosz1996“Equilibrium Exchange Rates in Transition Economies,”IMF Staff PapersVol. 44 pp. 430-61

    • Search Google Scholar
    • Export Citation

    HinkleLawrence E. and FabienNsengiyumva1999“External Real Exchange Rates: Purchasing Power Parity, the Mundell-Fleming Model, and Competitiveness in Traded Goods,” in: Lawrence E.Hinkle and Peter J.Montiel (eds.) “Exchange Rate Misalignment – Concepts and Measurement for Developing Countries,”Oxford University Press

    • Search Google Scholar
    • Export Citation

    International Monetary Fund (1997) “Trade Liberalization in Fund-Supported Programs”, EBS/97/163 (Washington).

    International Monetary Fund (1998) “Growth Experience in Transition Economies”, SM/98/228 (Washington).

    International Monetary Fund (1999) “The Baltics—Exchange Rate Regimes and External Sustainability”, SM/99/282 (Washington).

    KrajnyákKornelia and JerominZettelmeyer1997“Competitiveness in Transition Economies: What Scope for Further Appreciation?,”IMF Staff PapersVol. 45 pp. 309-362

    • Search Google Scholar
    • Export Citation

    LipschitzLeslie and DonoghMcDonald1992“Real Exchange Rates and Competitiveness: A Clarification of Concepts and Some Measures for Europe,”EmpiricaVol. 19

    • Search Google Scholar
    • Export Citation

    MarshIan and StephenTokarick1994“Competitiveness Indicators: A Theoretical and Empirical Assessment,”IMF WP/94/29

    MontielPeter1999“Determinants of the Long-Run Equilibrium Real Exchange Rate: An Analytical Model,” in: in: Lawrence E.Hinkle and Peter J.Montiel (eds.) “Exchange Rate Misalignment – Concepts and Measurement for Developing Countries,”Oxford University Press

    • Search Google Scholar
    • Export Citation

    RotherPhillipp2000“The Impact of Productivity Differentials on Inflation and the Real Exchange Rate: An Estimation of the Balassa-Samuelson Effect,”mimeo, IMF

    • Search Google Scholar
    • Export Citation

    SalterW.1959“Internal and External Balance: The Role of Price and Expenditure Effects,”Economic Record 35 pp. 226-38

    SamuelsonPaul A.1964“Theoretical Notes on Trade Problems,”Review of Economics and Statistics 46

    SwanTrevor1960“Economic Control in a Dependent Economy,”Economic Record 36 pp. 51-66

    World Economic Forum (1999) “Global Competitiveness Report”.


Prepared by Angana Banerji and R. Gaston Gelos


This paragraph is partly based on Hinkle and Nsengiyumva (1999). See also De Gregorio, Giovannini and Krueger (1994). For a more detailed discussion of the relation between different competitiveness indicators, see Lipschitz and McDonald (1992) and Marsh and Tokarick (1994).


This is often referred to as the “Australian” or “dependent economy” model, after Salter (1959) and Swan (1960).


For a discussion of these issues, see also EBRD (1997).


For a further discussion of these problems, see Krajnyák and Zettelmeyer (1998).


Two noteworthy attempts to overcome these problems have been made in the literature. Halpern and Wyplosz (1997) and Krajnyák and Zettelmeyer (1998) estimated equilibrium wages, which, under certain assumptions, can be linked to equilibrium real exchange rates. The latter article will be discussed further below.


The overall pattern of the REER based on producer price indices is similar, but this indicator shows a higher appreciation for the same period (27 percent). Excluding CIS countries—countries whose importance as trading partners is diminishing—from the computation of the REER did not have a noticeable impact on the indicator.


This index is a weighted average of external and internal liberalization and private sector conditions indices, and it has been constructed for transition economies for the period 1989–1995. The EBRD constructs similar indices. However, they are only available since 1994. For 1996 and 1997, the Denizer, De Melo, and Gelb index was updated based on these EBRD indices following Garibaldi, Mora, Sahay, and Zettelmeyer (1999).


Obviously, this is only one of many possible comparisons. In most transition economies, the REER followed a similar pattern since 1989: an initial sharp decline was followed by a steady appreciation (see Halpern and Wyplosz, 1997). Comparing the appreciation of the REER’s since their respective historical minima puts Bulgaria in the upper range.


See Krajnyák and Zettelmeyer (1998).


See Marsh and Tokarick (1994). The authors also provide a formal derivation of the link between the REER based on CPFs and unit labor costs.


Owing to the aforementioned lack good private sector wage data, the analysis below is based on public sector industrial wages.


Owing to lack of reliable data, labor shares were assumed to be constant. As a proxy for the relative price of nontradables, the same sectoral deflators as above are employed. Since the necessary capital stock data to compute total factor productivity is unavailable, labor productivities are used instead. This approximation is justifiable as long as the capital-labor ratios do not change much faster in one sector than in the other.


Typically, as transition economies develop, the traditional reliance on exports of heavy industrial goods and machinery is replaced by exports of new products with increased specialization at a finer level of product detail. See Growth Experience in Transition Economies, SM/98/228.


Bulgaria has made great strides in liberalizing its trade regime in recent years, and there are plans to undertake additional measures to liberalize trade during 2000. If these measures are fully implemented, Bulgaria’s trade regime will merit a 2 (“open”) as measured by the IMF’s index of aggregate trade restrictiveness, as compared to a 6 (“moderately restrictive”) in 1998.


Indeed countries which have made the biggest strides in liberalizing trade regimes are also those which seem to have experienced the most significant improvements in exports since the beginning of stabilization (Table 5 and Figure 16) as defined in Fischer, Sahay, Vegh (1996). This definition of stabilization is different from that used in Section B.


In other transition economies (for example, Estonia, Latvia, Lithuania) a rapid growth in exports has been linked to the transfer of technology, management, marketing skills, and corporate governance associated with strategic foreign investors (The Baltics—Exchange Rate Regimes and External Sustainability, SM/99/282).

Other Resources Citing This Publication