This Selected Issues paper examines the external sector issues in Hungary. It looks at some traditional indicators of external competitiveness, and provides basic background on key macroeconomic developments, including the current account and its financing. The paper examines various measures of the real effective exchange rate, the labor market considerations, and actual export performance. The paper concludes that although external competitiveness of the Hungarian economy has not been permanently impaired and the country is doing well from various vantage points, a significant dent has been put in competitiveness.

Abstract

This Selected Issues paper examines the external sector issues in Hungary. It looks at some traditional indicators of external competitiveness, and provides basic background on key macroeconomic developments, including the current account and its financing. The paper examines various measures of the real effective exchange rate, the labor market considerations, and actual export performance. The paper concludes that although external competitiveness of the Hungarian economy has not been permanently impaired and the country is doing well from various vantage points, a significant dent has been put in competitiveness.

I. External Sector Issues in Hungary1

A. Introduction

1. While transition economies like Hungary can be expected to run sizable current account deficits, the associated large capital inflows can pose risks. Hungary’s potential for rapid productivity growth provides investment opportunities well in excess of domestic savings. But foreign investors can lose confidence in a country running a high current account deficit, possibly leading to financial turbulence; and a major correction of the real exchange rate might then be necessary. This leads to the issue of the size of current account deficits appropriate for Hungary.

2. A number of methods seem relevant for assessing the size of current account deficits. While the results from using these methods must be interpreted with considerable caution, not least because of the application to a transition economy, they provide benchmarks which may at least flag concerns about the current account and the level of the real exchange rate. These benchmarks include those developed from:

  • a simple growth accounting framework to gauge the demand for foreign capital (and, thus, the required current account deficits) needed to achieve real convergence to EU income levels within a certain period;

  • an empirical estimate of the current account position compatible with model-based predictions of FDI inflows (as implied by cross-country work that draws on the combined experience of a large set of countries) and a stable ratio of external debt;

  • an application of recent panel estimates of the determinants of the current account balances to the case of Hungary;

  • a calculation of the underlying current account deficit based on a simple trade-equation approach—that is, the current account deficit in the absence of cyclical output gaps and after taking into account the lagged effects of (recent) real exchange rate changes, and

  • in the spirit of the IMF’s CGER analysis, an estimate of the change in the real exchange rate that would bring the current account from its underlying level to that in line with the benchmarks of the current account under the simplified assumption that all the adjustment would take place through the real exchange rate.

3. Before turning to these benchmarks, the paper looks at some traditional indicators of external competitiveness. For context, it provides basic background on key macroeconomic developments, including the current account and its financing. Then, it turns to various measures of the real effective exchange rate, along with analysis of labor market considerations and actual export performance.

B. Stylized Facts and Competitiveness Indicators

Growth, Inflation, and BOP developments in perspective

4. Hungary’s real growth performance early in the transition process disappointed compared to other Central European countries (CEC).2 The initial transition shock in the early 1990s led to a relatively steep decline in Hungarian real GDP exceeding that of other CECs (dubbed the CEC4 countries in what follows). Real growth rarely exceeded that prevailing in the CEC4 or in the euro area (Figure 1).3

Figure 1.
Figure 1.

Hungary, the CEC4, and the Euro Area: Real Growth and Inflation, 1990-2002 1/

(Percent)

Citation: IMF Staff Country Reports 2003, 125; 10.5089/9781451817928.002.A001

Sources: WEO; staff estimates.1/ CEC4 inflation data starting in 1991. Data for 2002 preliminary.2/ CEC4 data are (weighted) averages for the Czech Republic, the Slovak Republic, Slovenia, and Poland. The weights are based on 2001 GDP data, using market exchange rates. Countries in alphabetical order.3/ CEC4 1990-93 excluding Slovenia. CEC4 inflation in 1991 (not shown) is about 61 percent.

5. But, based, in part, on strong FDI inflows and export success, Hungary excelled in the second half of the 1990s. Real activity started expanding at a significantly higher rate after the mid-1990s. With real GDP growing at an estimated annual rate of about 4.4 percent between 1997 and 2002, Hungary stood out both vis-à-vis the other CEC countries (3.2 percent) and the euro area (2.3 percent). Among other factors, this growth spurt was due to the massive inflows of FDI in the wake of Hungary’s liberalization and privatization efforts in the preceding years (Figure 2, panel D).4 The vast majority of foreign investment was export-oriented, helping to increase net-exports (Figure 2, panel C) and, ultimately, real growth.

Figure 2.
Figure 2.

Hungary and the CEC4: Selected BOP Data, 1990-2002 1/

(Percent of GDP)

Citation: IMF Staff Country Reports 2003, 125; 10.5089/9781451817928.002.A001

Sources: IFS, WEO, and staff, estimates.1/ The Hungarian current account data before 1994 is settlement-based. All 2002 data are preliminary.2/ CEC4 data are (weighted) averages for the Czech Republic, the Slovak Republic, Slovenia, and Poland. The weights are based on 2001 GDP data, using market exchange rates. Countries in alphabetical order.

6. Inflation, too, showed a distinct behavior. While Hungary was spared the episodes of three-digit inflation haunting Poland in 1990 and Slovenia before 1994, inflation remained stubbornly high throughout most of the period, only falling to levels fairly close to the euro area average as late as 2002. CEC4 countries, on average, enjoyed significantly lower rates of inflation than Hungary.5

7. But perhaps the most striking difference between Hungary and the CEC4 was the development of the balance-of-payments (Figure 2):

  • While the CEC4 showed broadly balanced current accounts throughout the first half of the 1990s, Hungary recorded deficit levels exceeding 10 percent of GDP toward the end of the period (Figure 2, Panel A).6 Driven mostly by an improved balance in goods and services trade (Figure 2, Panel B), but partly reflecting fiscal consolidation, the current account deficit stabilized at more moderate levels between 1996 and 2001—broadly in line with developments in the CEC4 group.7

  • In 2002, after a deficit of only about 3½ percent of GDP in 2001, the current account deficit widened again, reaching over 4 percent of GDP. Although reinvested earnings are not yet included in the official statistics for Hungary’s current account and BOP because the data are unavailable, their inclusion could perhaps add 2-3 percentage points of GDP to the current account deficit, possibly bringing the 2002 deficit to 6-7 percent of GDP.8

  • Financing flows broadly followed the developments of the current account, reflecting both Hungary’s relative early privatization effort and its attractiveness as a destination for “green field investments” from abroad. The current account deficits of the first half of 1990s where mainly financed by large amounts of net portfolio investments and (as mentioned above) FDI inflows9—with the former peaking at about 10 percent of GDP in 1993 and the latter reaching similar heights in 1995—that were otherwise absent in the CEC area (Figure 2, Panel D and E).

  • In more recent years, by comparison, with privatization receipts fading, FDI into Hungary seems to have lost some momentum, stabilizing at levels around of 4½ percent of GDP in the 1996-2001 period before dropping to just about 1½ percent of GDP in 2002.10 At the same time, average FDI flows into the CEC4 group showed an upward trend, reaching an average close to 5 percent of GDP in 2002. Only part of this difference can be explained by the fact that, in contrast to the other CECs and most OECD countries for that matter (MNB 2002b), reinvested earnings are currently not taken into account when calculating FDI flows.

Competitiveness Indicators

Developments in Nominal and Real Effective Exchange Rates (REER)

8. The notable appreciation of the nominal effective exchange rate starting in late 2000 marks a structural break in its behavior over the last decade (Figure 3).11

Figure 3.
Figure 3.

Hungary: Effective Exchange Rate Measures, 1990-2002 1/

(1995-100)

Citation: IMF Staff Country Reports 2003, 125; 10.5089/9781451817928.002.A001

Sources: MNB; staff calculations.1/ Quarterly data. 2002 data preliminary or estimated.

9. Partoftheforint’s newfound strength reflects a change in the monetary framework. With the switch to a wider exchange rate band in May 2001 (to +15 percent around a central parity against the euro), the MNB’s room to appreciate the currency increased as it rejuvenated its disinflation efforts (Figure 1) and foreign capital continued to flow into Hungary (Figure 2).

10. Real exchange rate appreciation preceded the nominal rise. Reflecting mainly the positive inflation differential between Hungary and its trading partners, CPI- and PPI-based REER measures showed an upward trend as early as 1998 (Figure 3). The nominal appreciation of the forint only slightly amplified their upward momentum.12

11. However, an increase in CPI- or PPI-based REER does not necessarily imply a loss in competitiveness. Real convergence entails higher real growth rates, based, among other things, on high productivity growth in the tradables sector. As Balassa (1964) and Samuelson (1964) have shown, this might induce higher overall inflation: if productivity growth in non-tradables lags tradables but wages move along similar lines, the resulting price increase in non-tradables will increase overall inflation.

12. While hard to evaluate, it would seem plausible that a significant part of the observed real appreciation in CPI- or PPI-based REER was due to equilibrium effects. Estimates of the Balassa-Samuelson effect for different transition countries vary widely, suggesting that the resulting real appreciation could be anywhere in the range between 1 and 4 percent per annum (though more recent estimates, perhaps because of the advanced stage of structural reforms, tend to point to the lower end of the spectrum for Hungary).13 This would imply that the effect could explain between 20 and 50 percent of the observed average annual real appreciation over the 1998-2002 period of about 5 percent.

13. Mostly driven by wage increases, the REER based on unit labor costs (ULC)—arguably the most important determinant of international competitiveness14—followed a distinctly different pattern, however (Figure 3).15

  • The series declined throughout the second half of the 1990s, but the decline seems to have lost some momentum with the onset of disinflation.

  • The significant wage increases of 2001 and 2002, set against a background of a slowdown in productivity growth, were a significant contributor to the steep appreciation in the ULC-based REER that stands out in Figure 3. While the upward movement of the ULC-based REER sets in at approximately the same time as the appreciation of the nominal effective exchange rate, the increase of the ULC-based REER rate is significantly stronger than that of the NEER.

  • The increase in Hungarian unit labor costs relative to foreign unit labor costs explains about two-thirds and effective nominal appreciation the remaining one-third of the appreciation of the ULC-based REER during January 2001 to September 2002. During this period, the NEER appreciated by about 7.8 percent; Hungarian unit labor costs increased by about 19.4 percent, compared with an increase in foreign unit labor costs of just about 4.1 percent.

Labor market considerations

14. It is hard to blame the recent loss in competitiveness on labor market inflexibility. While it is difficult to encapsulate the impact of the multitude of institutional and legal factors that define labor market flexibility into a single measure, most observers would probably agree that Hungary’s labor market is among the least constrained in continental Europe.16 For instance, standard measures of employment protection (thought as inhibiting job creation by many economists) show Hungary leading both the CEC5 and the EU in deregulation (Table 1). At the same time, the replacement ratio in unemployment insurance (often interpreted as an important determinant of employees’ bargaining position in wage negotiations) exceeds both the EU and the CEC4 average.

Table 1.

Hungary and Selected Countries: Labor Market Regulation Indicators, 2002

article image
Sources: Riboud and others (2002), staff calculations.

Weighted average, using 2001 GDP data (at market exchange rates) as weights.

15. This positive assessment of labor market flexibility also finds support in assessments of overall Hungarian competitiveness. For instance, the World Economic Forum’s competitiveness rankings—taking into account both the macroeconomic framework and institutional quality at the micro level—consistently place Hungary ahead of most other transition and EU accession countries and among the top-30 countries overall.17 Hungary is also ranked first among the CEC countries in the EIU Report on World Investment Prospects.

16. But, in comparison with the EU average and some of the CEC4, Hungary’s labor taxes stand out as extraordinarily high (Figure 4). Income taxes plus employer’s social security contributions sum up to well above 50 percent of labor costs in Hungary, and are—on average—about 10 percentage points above those in the EU or, for instance, Poland. In fact, among current EU members, only Belgium’s labor taxes exceed the Hungarian tax level.

Figure 4.
Figure 4.

Hungry and Selected Countries Tax Wedge on Labor, 1993-2001

(Percent of ubor costs)

Citation: IMF Staff Country Reports 2003, 125; 10.5089/9781451817928.002.A001

Source OECD; staff calculations.
Hungary’s export performance

17. The picture emerging from the review of competitiveness indicators so far is mixed. On the one hand, levels of ULC-based REER measures are still rather low, both along the time dimension and compared to trade partners and competitors, suggesting a high level of competitiveness (see Figure 5). The same can be said for the relative flexibility of Hungarian labor market institutions. On the other hand, the tax burden on labor seems to be high by international standards and both the recent appreciation of the forint and the stark increase in labor costs suggest that Hungary has lost some of its competitive edge in recent years.

Figure 5.
Figure 5.

Hungary and Selected CEC Countries: ULC-Based REER, 1993-2002 1/

(1993 = 100)

Citation: IMF Staff Country Reports 2003, 125; 10.5089/9781451817928.002.A001

Sources: IMF; staff calculations.1/ Quarterly data. 2002 data preliminary or estimated.

18. Turning to actual export performance, it was strong vis-à-vis the EU through 2002. The EU is Hungary’s single most important export market, absorbing an estimated 75 percent of Hungary’s goods and services exports in 2002. During the 1990-2001 period, its share in overall EU imports more than tripled to more than 1 percent; and the available data for 2002 suggest a further increase to about 1.1 percent (Figure 6).

Figure 6.
Figure 6.

Hungary and CEC4: Share in EU Imports, 1993-2002 1/

(percent)

Citation: IMF Staff Country Reports 2003, 125; 10.5089/9781451817928.002.A001

Sources: IMF; staff calculations.1/ 2002 data is January-September.2/ CEC4 data are (weighted) averages forthe Czech Republic, the Slovak Republic, Slovenia, and Poland. The weights are based on 2001 GDP data, using market exchange rates.

19. Over time, Hungary was more successful than the average CEC Country in penetrating the EU market. While the CEC4’s average EU import share exceeded that of Hungary in the early 1990s and continued to increase, it did so at a slower pace than for Hungary (Figure 6). By 1997, Hungary had surpassed the CEC4 average.18 A comparison on a country-by-country basis leads to a somewhat less pronouncet picture, but Hungary’s dynamic export performance still was impressive.

20. More recently, however, this success seems to have come at the price of diminishing profit margins in the export sector. While export prices outgrew unit labor costs in manufacturing during the late 1990s, recent developments suggest a reversal of these trends (Figure 7). In 2001, as highlighted by the MNB (2002a) Report on Financial Stability, export sector profits declined not only in absolute terms but also relative to other sectors. Having enjoyed above-average profit margins throughout the second half of the 1990s, lower foreign demand and the appreciation of the forint added pressure from the revenue side as wage growth increased labor costs.

Figure 7.
Figure 7.

Hungary: Profits in the Export Sector, 1995-2002 1/

(1995=100)

Citation: IMF Staff Country Reports 2003, 125; 10.5089/9781451817928.002.A001

Sources: IMF; staff calculations.1/ Quarterly data. 2002 data preliminary.2/ Manufacturing.

21. Will Hungary’s success story continue? Many factors contributed to Hungary’s trade achievements during the last decade, including structural reforms, the inflow of export-oriented FDI, and, importantly, the gains in ULC-based competitiveness and the continued real depreciation of the effective exchange rate in the course of the 1990s. A corollary of the latter observation is that, conversely, the recent and significant dent into competitiveness, were it to expand, could pose problems for the export sector and the attractiveness of the country as a destination for FDI.

C. Benchmarks for the Current Account

22. This section looks at the size of the current account from several angles. First, it assesses the compatibility of different levels of the current account deficit with economic growth under different conditions represented by a growth accounting framework. Second, it evaluates the current account from the standpoint of debt dynamics. Finally, the determinants of the current accounts of a large sample of countries based on historical experience are used to assess the current account deficit for Hungary.

The current account and economic growth

23. Foreign savings channeled into domestic investment are an important factor behind real economic growth and income convergence. As a rule, a transition economy converging to the much higher EU income levels can be expected to exhibit a significant demand for foreign savings. At the same time, the relative scarcity of capital and the higher productivity compared to more developed economies should ensure a sufficient supply of foreign capital, not least because political risks are low and EU accession ensures compatibility of regulations and laws.

24. A simple growth model can help gauge Hungary’s demand for foreign savings and the implied current account deficits. Appendix I describes the theoretical underpinnings and data requirements of the exercise.19 The model builds on a number of assumptions. For instance, the capital-output ratio is kept constant and the economy is treated as being in a unique steady state in each year for which the demand is calculated. While restrictive, these assumptions allow benchmark results to be established in circumstances in which it would certainly be difficult to estimate more involved models. Recognizing the uncertainties surrounding such exercises, sensitivity analysis to changes in the underlying assumptions is undertaken.

Total factor productivity

25. The rate of growth of total factor productivity (TFP) is determined on the basis of an empirical relationship that depends on the average year of schooling and the income gap vis-à-vis the United States. The approach follows Doyle and others (2002), using estimates by Benhabib and Spiegel (1994).20 For Hungary, this results in a predicted TFP growth rate of 2.35 percent—a result close to the assumptions made, for instance, by the OECD (2000) and Darvas and Simon (2000). Based on standard output elasticity assumptions (see Appendix I), this translates into a contribution to GDP growth of about 3.5 percent per annum.

Estimates of the capital stock and depreciation rate

26. Estimates of the capital stock—more art than science in the best of circumstances—are notoriously difficult in the case transition economies. The very nature of transition from plan to market implied a dramatic re-evaluation of the existing capital stock, and researchers have made vastly different assumptions about the economic value of Hungary’s inherited private and public assets after 1990. As a result, estimates of the current capital stock differ as well. For instance, Hviding (1998) estimates a capital-to-GDP ratio (private and public capital, including residential construction) for the year 1997 around 1.8, while Doyle and others (2002) use a ratio of above 2 for 1999. By comparison, the U.S. Bureau of Economic Analysis estimates the capital-to-GDP ratio in the United States at about 2.7 in 1997 and about 2.8 in 2000.21

27. Some of the same difficulties pertain to the depreciation rate. The presence of pre-transition capital could lead to an overstatement of the effective capital stock, suggesting depreciation rates higher than, for instance, the annual 7 percent estimated for the United States.22 On the other hand, much of these capital goods will have been retired since 1990. In the absence of reliable estimates, a pragmatic approach is to assume a moderate depreciation rate of 7.5 percent as a baseline and provide alternative scenarios as a sensitive check.

28. A simple back-of the-envelope calculation places the capital-GDP ratio at around 2.4. Darvas and Simon (2000), in one of the more authoritative studies on the subject, establish a capital-to-GDP ratio excluding residential construction of about 1.7 in 1997.23 Assuming that residential construction amounts to about 30 percent of the overall capital stock24, this translates into a ratio of about 2.3. Based on actual investment data and an assumed average annual depreciation rate of 7.5 percent, this ratio would have increased slightly to about 2.4 by the year 2002, the starting point of the simulation exercise below.

Prospects for domestic savings

29. The prospects for domestic savings are difficult to evaluate. Hungary’s saving rate (as measured in the national accounts) has been volatile, starting at values as low as 11.5 percent of GDP in the early 1990s, reaching 25 percent of GDP in 1997, and, more recently, returning to levels of about 21 percent (Figure 8). Moreover, international experience does not necessarily provide a clear indication of what to expect. While the average savings rate in OECD countries decreased significantly from about 25 percent in the mid-1960s to levels less than 20 percent in the 1990s, savings rates in Eastern Asian emerging markets doubled between the mid-1980s and the mid-1990s to reach levels of about 30 percent. Like for other transition countries, the question arises which direction Hungary might follow.25

Figure 8.
Figure 8.

Hungary: Gross National Savings Rate, 1933-2002 1/

(Percent of GDP)

Citation: IMF Staff Country Reports 2003, 125; 10.5089/9781451817928.002.A001

Sources: IMF; staff calculations.1/ National accounts concept. Data for 2002 preliminary.

30. The savings rate is likely to increase modestly in the time ahead. Planned fiscal consolidation could raise public savings, but part of this effect might be offset by a (Ricardian) reduction in private savings. Moreover, the Eastern Asian example of a strong positive relationship between growth and domestic savings might not necessarily apply to transition economies. A recent cross-country analysis of the determinants of savings in transition countries, including Hungary, even found a significant negative correlation between growth and domestic savings for the 1992-99 period26—perhaps due to the important role of foreign savings for the financing of investment. In what follows, it is assumed that national savings gradually converge to a level around 22.5 percent of GDP by the second half of the decade.

Labor supply

31. Employment growth is likely to turn negative in the long run.27 In addition, a gradual decrease in the unemployment rate, currently at 5.9 percent, to 5 percent by 2020 is assumed. This is consistent with employment growth averaging about 1 percent in the 2003-10 period, before falling by about 0.3 percent a year in the 2010s, and by about 1 percent over the long term, reflecting Hungary’s rapidly aging population.28

Simulation results for the current account

32. Given these assumptions, the growth model can be put to work to provide insights into the links between economic growth and the current account deficit. Assuming that the economy retains a constant capital-GDP ratio and savings rate, the demand for foreign savings is determined by TFP growth, employment growth, and the depreciation rate on the capital stock (as in Appendix I). Table 2 depicts the implied current account deficit for alternative rates of capital depreciation between 6 and 9 percent.

Table 2.

Hungary: Current Account Simulations with Constant Capital-GDP Ratios

(Period-Average in Percent of GDP), 2003–50

article image

The period 2003-10 will be referred to as “medium term” in what follows.

33 The main results can be summarized as follows:

  • Real GDP grows by about 4 percent over the medium term before gradually converging to its lower long-run level around 2.5 percent. This reflects the (constant) contribution of TFP to growth and the declining supply of labor.

  • Assuming real per capita GDP growth of 2 percent in the current EU area, this would imply that Hungary’s GDP per capita would reach 90 percent of the EU average by 2040 and full income convergence in 2050.

  • Under the baseline assumption of a 7.5 percent annual depreciation, current account deficits amount to about 6.5 percent of GDP in the medium run. Mostly because of the underlying assumptions on employment growth, the model predicts a downward trend in the deficit figure in the longer run.

  • The current account deficit levels change with the underlying depreciation rate. A lower rate of capital depreciation of 6 percent would imply a current account deficit of about 3.4 percent of GDP in the medium run. In contrast, a higher depreciation rate of 9 percent would result in a significantly larger medium-term deficit of about 9.5 percent.

  • Lowering the capital-to-GDP ratio from 2.4 to 2.0 or increasing it to 2.8 has a similar strong effect on the implied current account deficit. At a deprecation rate of 7.5 percent, the implied medium-term deficit could be as low as 2.4 percent of GDP or as high as 10.5 percent of GDP depending on whether the lower or the higher ratio is selected.29

  • Real growth and, thus, Hungary’s demand for foreign savings also depend on the assumed contribution of TFP. For instance, raising projected TFP growth from the 2.35 to 2.5 would increase the average current account deficit during the simulation period by about ½ a percentage point. Similarly, a lower-than-expected saving rate would lead to larger forecasted current account deficits throughout the forecasting period.

34. A possible interpretation of these results is that Hungary’s demand for foreign savings could be compatible with moderate current account deficits in the medium and long run. Based on the baseline scenario, a plausible benchmark for current account deficits that would not unduly constrain growth would be in the general vicinity of 6.5 percent of GDP in the medium run.

FDI and net external debt

35. This section assesses the level of the current account deficit from the perspective of debt dynamics. This requires an analysis of FDI, to shed light on the possible split between debt and non-debt financing.

36. High inflows of foreign equity, initially based on privatization and more recently on other forms of FDI, allowed a significant reduction of Hungary’s net foreign debt levels, fostering sustainability.30 While net external liabilities remained fairly stable at around 60 percent Of GDP after the mid-1990s (Figure 9), their composition changed, with equity substituting for debt-based financing. Arguably, this development contributed to reducing risks of financial distress in case of adverse external shocks, as equity is often less liquid in case of a loss of confidence. Moreover, the accumulation of FDI adds to the capital stock, tends to involve the transfer of know-how and human capital, thereby supporting growth and strengthening the overall health of the economy. Finally, a larger FDI stock may help reduce problems of asymmetric information for foreign investors, providing incentives for further equity inflows.31

Figure 9.
Figure 9.

Hungary: Net External Liabilities, 1990-2002 1/

(Percent of GDP)

Citation: IMF Staff Country Reports 2003, 125; 10.5089/9781451817928.002.A001

Sources: MNB; IMF; staff caluclations.1/ 2002 preliminary.

37. With EU accession almost assured, Hungary’s international financial standing improved further. While most accession countries saw their overall sovereign rating for long-term foreign currency debt upgraded since 1998—including the recent increase related to the confirmation of accession32—Hungary’s ascent was relatively swift (Table 3). In November 2002, Moody’s graded Hungary similar to Greece, and only Slovenia enjoys a higher ranking.33

Table 3.

Hungary and Selected Countries: Moody’s Ranking, 1998 and 2002

article image
Sources: Moody’s; staff calculations.

Speculative Grade.

Increase in rating ranks between November 1998 and November 2002.

38. However, the reduction in the net foreign debt ratio to GDP slowed down in recent years and, without corrective action, could be reversed in the wake of debt policy changes and higher fiscal deficits. The authorities have resumed foreign borrowing at a point when the general government deficit reached an estimated 9.5 percent of GDP (2002, on an ESA-95 basis). But, if the authorities’ target is met, the deficit would fall to 4.5 percent this year, helping to reduce the possibility of unfavorable external debt dynamics.

Hungarian FDI performance in perspective

39. A continued inflow of FDI could significantly extend the scope for current account deficits from a debt perspective. As Figure 10 illustrates, since the mid-1990s, FDI equity investments in Hungary were, on average, significantly larger (and, in addition, somewhat more steady) than FDI non-equity and portfolio equity 34 Moreover, for all but the most recent years, FDI non-equity investment flows, mostly intercompany loans of multinationals operating in Hungary, moved more or less in line with FDI equity investments. This suggests that at least part of the non-equity FDI inflows could be based on longer-term financial commitments akin to FDI equity inflows.

Figure 10.
Figure 10.

Hungary: FDI and Portfolio Equity Inflows, 1996-2002 1/

(Percent of GDP)

Citation: IMF Staff Country Reports 2003, 125; 10.5089/9781451817928.002.A001

Sources: MNB; IMF; staff caluclations.1/ 2002 preliminary.2/ Mostly intercompany loans.

40. The level of FDI inflows during the 1998-2001 period exceeded those into countries with similar financial ratings and income levels, but seems to lag behind the other CEC countries (Table 4). However, as already mentioned, the Hungarian BOP statistics do not include reinvested foreign earnings. Assuming reinvested earnings amounted to about to 2-3 percent of GDP and including such figure in Table 4 would have placed Hungary ahead of the other CECs.

Table 4.

Hungary and Selected Countries: Reported Average FDI Inflows, 1998-2001 1/

article image
Sources: IFS, Moody’s Ranking (as of January 2003); staff calculations.

The time horizon is chosen to match the econometric exercise below (see paragraph 42 and Appendix II).

Excluding reinvested earnings.

41. A disadvantage of comparisons based on (ex post) outcomes is that they fail to control for differences in country-specific determinants of FDI. For instance, countries might differ in demand for foreign savings, their capacity to absorb investments from abroad, and their attractiveness for such capital flows based on factors such as political stability.35 A case in point is the high ranking Hungary receives based on the U.N. (2002) “FDI potential index” that, among other things, incorporates information on GDP per capita, infrastructure and human capital, R&D expenditure, human capital, and country risk. Out of 141 countries, Hungary ranks at 41—right among the majority of accession countries.36

42. A more extensive empirical analysis of the historical experience of a large set of countries suggests that Hungary’s potential to attract high levels of FDI inflows remains strong. Appendix II describes a set of pooled regression models explaining average FDI inflows in percent of GDP in a sample of 96 countries over the 1998-2000 period. The models relate FDI to a number of demand- and supply-side arguments, including the above mentioned U.N. potential FDI indicator. As expected, the latter is found to have a positive impact on FDI inflows. Other findings include:

  • Stock of FDI: A larger stock of past FDI investment increases current FDI inflows, but at decreasing speed. These points to decreasing-returns-to-scale, either based on decreasing real returns or the fact that foreign investors learn more about the qualities the destination country as they continue to invest.

  • Savings: Higher domestic savings reduce the demand for FDI inflows.

  • Inflation. Higher inflation rates deter foreign investors. The effect is somewhat stronger at moderate levels of inflation, that is, investors react more averse to an increase in inflation from, say, 5 to 15 percent than to an increase from 90 to 100 percent per year. A likely explanation is that investors believe they learn less about the stability of a possible FDI destination if the increase occurs at already very high levels of inflation.

  • Interest rate spread: FDI inflows increase (after controlling for inflation) as domestic interest rates rise relative to LIBOR.

43. Based on the historical experience captured by the empirical model, a baseline forecast for FDI inflows including reinvested earnings of a minimum of 5 percent of GDP would not be unwarranted. The point estimate is around 7.9 percent of GDP, with a sizable forecast standard error of about 2.9 percent of GDP. To err on the side of caution, the baseline has been set one standard error below the point estimate.

Current account implications of a stable debt-GDP ratio

44. The current account deficit that would stabilize the ratio of net foreign debt to GDP is the sum of these potential FDI inflows and the level of debt-creating inflows that would stabilize the debt-GDP ratio.37 Details are provided in Appendix III.

45. Given current debt levels, a plausible estimate of the debt-stabilizing current account deficit (including reinvested earnings) is around 6.5 percent to GDP in the baseline scenario (Table 5).

  • In the baseline scenario in Table 5, assuming real GDP growth of 4 percent and estimated FDI inflows of 5 percent of GDP, the current level of net foreign debt of 20-25 percent of GDP could be maintained with debt-creating capital inflows of 1.2-1.5 percent of GDP, implying current account deficits of 6.2-6.5 percent of GDP (see Panel A).

  • Under the more optimistic assumptions of the alternative scenario, holding real GDP growth constant at 4 percent, but assuming higher nominal GDP growth in foreign currency terms and higher FDI inflows of 7.9 percent of GDP, this range increases to 9.4-9.8 percent of GDP (see Panel B).

Table 5.

Hungary: Debt Stabilizing Current Account Deficits 1/

article image
Source: Staff calculations.

The shaded rows indicate current account deficits broadly consistent with current levels of external debt.

As discussed in Appendix III, the growth rate of nominal GDP in foreign currency terms can be decomposed in the sum of the change in the REER, the change foreign GDP deflator, and real GDP growth.

In both cases, lowering the assumed real growth rate of GDP from 4 to 2 percent would imply a lower debt-stabilizing current account deficit by about ½ percentage point of GDP. Equivalently, if real GDP growth were not 4 but 6 percent, the implied current account deficit would increase by about ½ percentage point.

The “Normal” current account balance

46. An alternative approach to finding a benchmark for the current account balance is to exploit the historical experience of other countries. The principle idea is to extract a measure of the "normal" or "equilibrium" current account balance that can be compared to actual current account balances (Isard and others 2001). While the approach might not fully take into account the particular characteristics of transition economies, it can nevertheless provide a rough idea of the orders of magnitude.

47. Such a benchmark can be computed based on a panel study by Chinn and Prasad (2000).38 They investigate the relationship between the current account deficits in a large sample of developing countries (excluding transition economies) in the period 1971-95 and economic factors that might influence both their saving-investment balance and their access to international capital markets. The relevant explanatory variables include:

  • Fiscal balance: The fiscal balance has the expected significant positive impact on the current account balance (in percent of GDP).

  • Financial deepening: The ratio of M2 over GDP is positively related to current account balances, suggesting that financial deepening decreases the necessity to seek foreign financing.

  • Old-age and young-age dependency ratios: Higher ratios decrease the current account balance, perhaps due to a decrease in the savings rate.

  • Openness: Higher degrees of openness (measured as the sum of exports and imports over GDP), associated with an enhanced capacity to generate exports or vary import demand in line with possible debt-payment constraints, decrease current account balances.

  • Terms of trade volatility: Larger swings in a country’s terms of trade increase the current account balance.

  • Relative GDP per capita: There is non-linear relationship between the relative per capita income (compared to the U.S. using PPP exchange rates) and the current account. While higher income at lower levels tends to imply more access to international markets and, thus, lower current account balances, this effect is reversed or reduced at higher income levels.

48. This approach, using a general government deficit of 3 percent of GDP in line with the authorities’ medium-term target, implies a “normal” current account balance (including reinvested earnings) of about 5.5 percent of GDP for Hungary.39

D. An Application of the CGER Model to Hungary 40

Short introduction to the model (and Its Caveats)

49. At the core of the CGER approach to evaluating the level of the real exchange rate is a comparison of the underlying current account with a benchmark. The underlying current account adjusts the actual balance by taking into account the effects of the business cycle and recent real exchange rate changes based on a simple partial equilibrium trade model (see below). The CGER approach compares this underlying current account balance with a benchmark for the level of the medium-term sustainable balance, labeled the saving-investment norm.

50. Over the medium run, any deviation of the underlying current account balance from the saving-investment norm is assumed to be countered by a change in the real exchange rate. For instance, an underlying current account deficit in excess of the norm, would imply a real depreciation and the size of the calculated depreciation would be an indicator of the degree of potential misalignment of the current real exchange rate. The mechanics of this exercise, that is, the mapping of exchange rate changes into changes in trade flows, is based on the same trade model and assumptions that are used to calculate the underlying current account balance.

51. The partial-equilibrium approach, while helpful in gauging some of the more important forces that may influence the current account, has its limitations. As 1 linkie and Montiel (1999) point out, the underlying structure of the CGER approach, a simple two-goods production model of the Mundell-Fleming type, precludes, among other things, the analysis of terms-of-trade shocks when calculating the medium-term current account balance. Even more importantly, the recursive partial equilibrium nature of the approach ignores other equilibrating forces: given the gap between the actual and the underlying current account, the CGER approach estimates the required change in the real exchange rate but keeps key macroeconomic variables, such as domestic investment and private and public saving, constant41 The absence of self-correcting general-equilibrium forces other than the real exchange rate could potentially introduce an upward bias in the required change in the real exchange rate. That being said, the simplicity of the CGER approach does help to evaluate some of the more crucial trade-related determinants of the current account in a transparent fashion, avoiding the “black box” characterizing many general equilibrium approaches.

52. Moreover, temporary factors reflected in the actual current account could bias the result. To the extent that the actual balance is influenced by negative one-off factors (which could, for instance, include high oil prices, negative shocks to tourism, or a temporary fiscal expansion) that would fade and thereby lead to smaller current account deficits without specific action being taken, there would also be an upward bias in the required change of the real exchange rate. The exercise below will provide some sensitivity checks along these lines.

53. Finally, as in any “applied” economic model, the CGER approach involves making a number of critical assumptions. However, there is, for example, hardly a consensus regarding the value of real trade elasticities in transition economies. As already mentioned, significant uncertainties also surround the choice of the proper current account benchmark. In what follows, different assumptions will be used to illustrate the sensitivity of the results with regard to these uncertainties.

What saving-investment norm?

54. No single current account benchmark is without drawbacks. Section C discussed a variety of benchmarks for current account balances for Hungary. The resulting baseline benchmarks range from about -5.5 percent of GDP to -6.5 percent of GDP, but a case could be made for somewhat higher values, for instance, if more optimistic assumptions on expected FDI flows were made.

55. In what follows, a benchmark current account balance around 6.5 percent of GDP will be assumed as baseline.

Underlying current account

56. The underlying current account is the balance that would emerge if all economies were operating at potential output at prevailing real exchange rates. The concept is used in the IMF’s CGER exercises for current account and exchange rate assessments in major industrial countries (see Isard and Faruqee 1998), but recently efforts have been made to extend the exercise to emerging markets and developing economies (Isard and others 2001).

Framework and assumptions

57. The concept is based on a standard partial equilibrium trade model, taking into account the real exchange rate and the state of the business cycle. Export volume depends on the current and lagged values of the REER and the trade-weighted output gap abroad, while imports depend on the current and lagged values of the REER and the gap between domestic GDP and potential output. A detailed explanation of the empirical application of this concept can be found in Isard and Faruqee (1998)—Appendix IV provides some technical details.

58. Applying the approach to Hungary’s requires a number of technical assumptions:

  • Real exchange rates. While no single REER indicator is best fit to explain trade fluctuations in all countries (Marsh and Tokarick 1994), the analysis in Section B supports using the ULC-based REER in the case of Hungary. The model described in Appendix IV allows for impact lags of up to three years, broadly in line with the findings of a number of empirical studies (Ghei and Pritchett 1999)42 Presuming wage moderation in line with government targets, solid productivity growth, and the absence of upward pressure on the nominal exchange rate vis-à-vis major trading partners, the working assumption here is that the ULC-based REER remains constant in 2003.

  • REER elasticities: Estimated elasticities even for a given REER measure vary widely across countries.43 In particular, trade flows in industrial countries are often found to react stronger to real exchange rate changes than exports and imports of developing countries, which, as a rule, face more volatile real exchange rate movements.44 There are indications that price elasticities in transition economies were even lower in the past.45 In light of these uncertainties, calculations of the underlying current account are performed across two different sets of elasticities.

  • Output gaps: Staff estimates place the output gap in Hungary in 2003 at around -2 percent of potential output. Reflecting mostly lower potential growth, the (trade-weighted) output gap for Hungary’s trade partners is assumed to be -1½ percent. Following the CGER approach, the elasticity of exports to a decrease in the foreign output gap is set at 1.14, the elasticity of imports to a decrease in the domestic output gap at 1.26. The discussion below sheds some light on the relative importance of these assumptions.

  • Current account. Staff projects a current account deficit (custom-based, but excluding reinvested earnings—see Section A) of about 4.8 percent in 2003. Some adjustments are made with regard to reinvested earnings and planned fiscal consolidation (described in the footnotes to Table 6 below, which reports the underlying current account using the CGER approach).

Table 6.

Hungary: Estimated Underlying Current Account Balance, 2003

article image
Source: Staff calculations.

Adjustment are made for (i) reinvested earnings, which increase the current account deficit by an assumed 2.5 percent of GDP, and (ii) the effects of meeting the government’s deficit? target of 4.5 percent of GDP in 2003 and 3 percent in 2004 by assuming a downward correction of the size of the current account deficit of about 1.5 percent of GDP.

Elasticities (1) and (2) are compatible with the CGER assumptions for developing and industrial countries, respectively (Isard and others 2001).

Results

59. Hungary’s underlying current account deficit, including reinvested earnings, could be much higher than its actual level. In column A, Table 6 reports estimates of the underlying current account balance under different assumptions on price elasticities. Changing from the lower elasticities of exports and imports in row (1) to the elasticities more characteristic of industrial countries in row (2) increases the underlying current account deficit.

60. The results are mostly indicative of the stark real appreciation in recent years, with business cycle effects playing a comparatively minor role. Closing the output gaps in the baseline scenario contributes less than 1 percentage point of GDP to the difference between the actual current account deficit and the underlying balance. Thus, the results would not change dramatically, if the adjustments for the business cycle were not made.

61. As might be expected, rebasing the calculations in the alternative scenario alters the picture. As a sensitivity check, column B in Table 6 presents results under the alternative assumption that the starting point for the current account deficit is only 2 percent of GDP—less than half the size of the deficit in the baseline scenario. The alternative scenario could be interpreted as taking into account transitory factors (other than fiscal policy) that might increase the observed actual current account in the short run.46 While relatively low, a figure of 2 percent is not outside historical limits.47 However, even in the alternative scenario, the underlying current account deficit could still reach an uncomfortably large size.

Is the real exchange rate overvalued?

62. Calculating the real exchange rate adjustment, using the CGER methodology, that would close the gap between the underlying current account deficit and the benchmark, raises concerns about the level of the real exchange rate (Table 7). In CGER terminology, significant deviations from equilibrium in excess of 15 percent, should be considered a warning flag, warranting serious consideration of a real misalignment (Isard and Faruqee 1998).

  • In the baseline scenario in column A, the computed REER adjustment exceeds the 15 percent threshold. A decrease in import and export elasticities tends to increase the computed adjustment. As mentioned earlier, empirical studies have found trade volumes in transition economies to be even less sensitive to exchange rate changes than in trade in many developing countries.48 However, a reasonable argument could also be made that the sensitivity of trade flows to real exchange rate changes increases over time as exchange rate volatility decreases.

  • Under the alternative scenario, the computed adjustment in the REER is much smaller.

  • Raising the benchmark for the size of the current account deficit to, say, 7.5 percent of GDP would change the result.49 At one extreme, it would lower the calculated REER adjustment from 37 to 29 percent (column A, row 1); and from 9 percent to 6 percent at the other (column B, row 2). The effect of lowering the benchmark to 5.5 percent of GDP would be of similar magnitude but in the opposite direction.

Table 7.

Hungary: REER Adjustment Suggested by the CGER Model, 2003 1/

article image
Source: Staff calculations.

Adjustments could reflect wages, productivity, and taxation, as well as the nominal exchange rate.

63. These results suggest that greater wage restraint would have many advantages. As previously discussed, wage increases explain roughly two-thirds of the steep acceleration of Hungary’s ULC-based REER in the past two years, making wage moderation key to preventing further (or to reversing past) real appreciation. Other available policy options affecting the REER include reducing the labor tax burden falling on employers (while being careful to maintain fiscal consolidation) and measures to foster productivity growth. Moving resolutely along these lines would have the advantage of lessening the adjustment burden falling on the nominal exchange rate and limiting the likelihood of a disorderly short-run correction.

E. Concluding Remarks

64. While the external competitiveness of the Hungarian economy has not been permanently impaired and the country is doing well from various vantage points, a significant dent has been put in competitiveness. This is reflected in the recent appreciation of the ULC-based real effective exchange rate, about two-thirds of which stems from wages and other labor market developments, and one third from nominal exchange rate appreciation.

65. Taking into account the lagged effects of the recent exchange rate appreciation, the current account deficit is a concern. While it is extremely difficult to quantify the precise effect of the appreciation of the exchange—not only because of the uncertainty over the magnitudes of the relevant elasticities but also because of the difficulties in sorting out the complex factors that affect the behavior of exports and imports—the various exercises undertaken in the paper at least raise warning flags that the external current account in Hungary could be moving into unwelcome territory if offsetting measures were not taken.

66. Notwithstanding the array of caveats and uncertainties associated with the various methods used in this paper, a current account deficit on the order of about 4½ percent of GDP would not seem especially problematic (on the basis of the official statistics that for now exclude reinvested earnings). It would likely be consistent with sustainability, in the sense of avoiding unfavorable debt dynamics. This is consistent with the earlier Fund study by Beaumont (1999). Moreover, over time, a deficit of this size would probably not unduly restrain economic growth, an important consideration in relation to real convergence with current EU countries.

67. This outcome for the current account implies the need for adjustment measures. One element in the adjustment process—as suggested, however imprecisely, by the CGER exercise—would be through future adjustments in the real effective exchange rate. This does not only include possible changes in the nominal exchange rate, but, probably more importantly, the need for wage restraint. Fiscal consolidation is also key, not only directly as a contributor to external adjustment but also by unburdening monetary policy and thereby helping to avoid unwanted nominal exchange rate appreciation. Of course, actions to raise productivity would also help, and could possibly be prompted by the recent pressures on profit margins in the tradable goods sector.

APPENDIX I: The Investment Ratio in a Long-Term Growth Scenario

Consider a simple economy where in any period t output or GDP (Y) is produced based on a standard Cobb-Douglas function with labor (L) and capital (K) as inputs:

Yt=AtLtαKt1α,(1)

where α and (1- α) are the elasticities of output with respect to employed labor and capital, respectively, and A captures total factor productivity (TFP). In the steady-state, the capital-output ratio is constant and, thus, GDP and the capital stock grow at the same rate. Therefore, the long-run rate of GDP growth can be written as

Y^t=A^tα+L^t,(2)

where the “ʌ” marks the growth rates of GDP, TFP, and employment, respectively. To calibrate the output elasticity with regard to TFP, a widely used assumption for a is 2/3, which is broadly in line with the wage share in GDP in many developed economies.

Per definition, the investment-to-GDP ratio in any period t is

ItYt=KtYt(1δ)Kt1Yt,(3)

with δ representing the depreciation rate. After rewriting

ItYt=KYδ+Y^1+Y^,(4)

where again use has been made of the fact that the capital-GDP ratio is constant in the steady state, i.e., that Kt/Yt = Kt-1/Yt-1 = K/Y. Thus, the investment ratio in period t depends on the capital-GDP ratio, the rate of GDP growth—defined in (2)—and the rate of capital depreciation.

APPENDIX II: Estimating Hungary’s FDI Potential

Table A1 presents the results from a set of pooled regression models explaining average FDI inflows in percent of GDP in a sample of 96 countries over the 1998-2000 period.50 The most extensive model explains about 56 percent of the variation across countries and is reasonably robust with regard to other specifications.

Table A1:

Explaining FDI in a Pooled Regression

article image
Sources: U.N.; IFS; World Bank; staff calculations.

White Heteroskedasticity-Consistent Standard Errors and Covariance.

Absolute t-Statistics in parenthesis.

The result show that the FDI determinants summarized by the U.N. (2002) indicator of FDI potential and the stock of existing FDI are among the more important explanatory variables. The model represented by the column labeled (1), which excludes other arguments, already explains about 45 percent of the variation in the sample. In particular:

  • FDI is the larger the higher the U.N. indicator for investment potential. The indicator summarizes, among other “classical” determinants of FDI, per capita GDP, infrastructure, and country risk (see Table A2 for details). All of them are commonly expected to influence both demand and supply of FDI across countries.

  • There is a positive but non-linear connection between the existing stock of FDI and current FDI flows. While the presence of past FDI investment is positively correlated with current FDI, its marginal impact declines as the stock increases.51 This result is compatible both with classical decreasing-returns-to-scale arguments and the concept of asymmetric information, i.e., the idea that foreign investors learn more about the qualities the destination country as they (continue to) invest more.

Table A2.

Data Definition and Sources54

article image

Additional variables seem to be robustly related to current FDI investment:

  • Savings: Higher domestic savings reduce the demand for foreign savings and, thus, also FDI.

  • Inflation: Higher inflation rates seem to deter foreign investors. While the impact is somewhat smaller at very high levels of inflation (most likely because the information content of another percentage increase at, say, 100 percent annual inflation, is limited), the net-impact of the squared and non-squared term is strictly negative across the sample.

  • Interest rate spread: FDI flows are increasing in the difference between domestic interest rates and LIBOR, suggesting perhaps that the spread signals differences in the potential (real52) productivity of investments.

  • Openness, a variable that could be expected to signal a country’s involvement in international trade and factor flows, is only marginally significant.53

As a robustness test, a number of other variables were introduced into the model. These include measures of capital controls and financial deepening and indicators of human capital, health, income distribution, and country size. None of these variables showed a robust correlation with FDI performance.

The last row in Table A1 reports the predicted level of FDI in percent of GDP for Hungary for each model, that is, the level implied by the respective estimated vector of coefficients. These forecasts represent an estimate of Hungary’s FDI potential based on the historical benchmark experience of all sample countries. The more complete models (2) to (4) predict FDI levels of about 7.9 percent of GDP, with a standard error of about 2.9.

Predicted inflows should be interpreted as including reinvested earnings. Since, as a rule, most country data within the sample includes reinvested earnings, the same applies to the model-based forecast of FDI inflows into Hungary. Excluding reinvested earnings Hungary’s FDI potential could be about 5-6 percent of GDP.

APPENDIX III: Current Account Deficits to Stabilize the Debt-GDP Ratio

The current account deficit (CAD*) that would stabilize the ratio of net foreign debt to GDP is the sum of these potential FDI inflows (FDI*) and the level of debt-creating inflows that would stabilize the debt-GDP ratio (DCI*). That is,

CAD*=FDI*+DCI*(1)

In discrete time, the condition for a constant ratio of net foreign debt (D) to nominal GDP (YnomF) denominated in foreign currency is

DtYnom,tFDt1Ynom,t1F=0,(2)

with Ynom,tF=et/Ynom,t, where e is the nominal exchange rate defined in terms of the currency composition of net foreign debt and Ynom stands for nominal GDP in national currency terms. Equation (1) implies that debt-creating capital inflows in percent of GDP that keep the net-foreign-debt-to-GDP constant in period t are

DCI*=DtDt1Ynom,tF|Dt/Yt=Dt1/Yt1=Dt1Ynom,t1FY^nom,F1+Y^nomF,(3)

where Ynom,tF/Ynom,t1F=Y^nomF is the growth rate of nominal GDP in foreign currency terms. Assuming that PPP holds, the latter can be decomposed into

Y^nom,F=Y^+e^real+π*,(4)

with Ŷ denoting real GDP growth, êreal the growth rate of the real exchange rate, and π* the growth rate of the foreign GDP deflator in terms of the currency composition of foreign debt.

Assuming medium-term GDP growth rates of 4 percent, foreign inflation of 1-2 percent, and, for illustration, a real (equilibrium) appreciation of about 1—4 percent, a medium-term range of Ŷ would be 7-11 percent.

APPENDIX IV: Underlying Current Account Deficits

Following Isard and Faruqee (1998), the trade-equation explaining the ratio of the current account balance to GDP, ca, is

cat=α[βmm+βxx][0.6ereal,t+0.25ereal,t1+0.15ereal,t2]+mθereal,t,mnmgapt+mnxgapt*(1)

where α is a constant, m and x are the ratio of exports and imports to GDP, gap and gap* are the domestic and the (trade-weighted) foreign output gap, and ereal, t represents the average level of the real exchange rate in year t. The second term in equation (1) models the impact of present and lagged levels of the effective real exchange rate on exports and imports, with a 3-year lag structure similar to the one used in the IMF’s Multimod model. The third term in (1) depicts the pass-though to import prices.

For θ = 1, the assumed pass-through would be immediate and full (as in Isard and Faruqee 1998), while values 0 < θ < 1 indicate a slower pass-through. Export prices are assumed to be independent of the exchange rate. The fourth and last term in (1) models the relation between the output gaps and ca. The terms βmx ղm, and ղx are partial elasticities.

Based on equation (1), the underlying current account cau can be found by assuming closed output gaps and a constant exchange rate, ereal, thought to prevail at the time of analysis:

cau,t=α[βmm+βxx]ereal+mθereal(2)

Subtracting (1) from (2) and rearranging yields

cau,t=cat[βmm+βxx][(erealereal,t)+0.4(ereal,tereal,t1)+0.15(ereal,t1ereal,t2)],+mθ(erealereal,t)+mηmgaptmηxgapt*(3)

In other words, the underlying current account can be calculated as the sum of the observed current account and a correction term, which depends on present and past changes of the real exchange rate and the domestic and foreign output gaps. Ceteris paribus, the sign of the correction term is more likely to be negative, if the real exchange rate follows a positive trend and domestic (but not foreign) output deviates significantly from potential.

References

  • Arvanitis, A., 2002, Determinants of Foreign Direct Investment in South Africa, IMF Staff Country Report No. 03/18 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Backé, P., and others, 2002, “Price Dynamics in Central and Eastern European EU Accession Countries,Österreichische Nationalbank Working Paper, No. 61.

    • Search Google Scholar
    • Export Citation
  • Balassa, B., 1964, “The Purchasing Power Parity Doctrine: A Reappraisal,Journal of Political Economy, Vol. 72, pp. 58496.

  • Beaumont, C., 1999, Sustainability in the Hungarian Balance of Payments, IMF Staff Country Report No. 99/27 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Beaumont, C., 1999, The Trade Balance and the Real Exchange Rate, IMF Staff Country Report No. 99/27 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Benhabib, J., and M. Spiegel, 1994, “The Role of Human Capital in Economic Development: Evidence From Cross-country Data,Journal of Monetary Economics, Vol. 34, pp. 43173.

    • Search Google Scholar
    • Export Citation
  • Carlson, A., and L. Hernández, 2002, “Determinants and Repercussions of the Composition of Capital Inflows,IMF Working Paper 02/86 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Chinn, M., and E. Prasad, 2000, “Medium-Term Determinants of Current Accounts in Industrial and Developing Countries: An Empirical Exploration,IMF Working Paper 00/46 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Coricelli, F., and B. Jazbec, 2001, “Real Exchange Rate Dynamics in Transition Economics,CEPR Discussion Paper, No. 2869.

  • Darvas, Z., and A. Simon, 2000, “Capital Stock and Economic Development in Hungary,Economics of Transition, Vol. 8, No. 1, pp. 197223.

    • Search Google Scholar
    • Export Citation
  • Daseking, C., 2001, “How Large Should Poland’s Current Account Deficit Be?”, IMF Selected Issues Paper SM/01/64 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • De Broeck, M., and T. Sloek, 2001, “Interpreting Real Exchange Rate Movements In Transition Countries,IMF Working Paper 01/56 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Dibooglu, S., and A. Kutan, 2000, “Sources of Real Exchange Rate Fluctuations in Transition Economies: The Case of Poland and Hungary,ZEI Working Paper, B14.

    • Search Google Scholar
    • Export Citation
  • Doyle, P., Jiang, G. and L. Kuijs, 2002, “Real Convergence to EU Income Levels: Central Europe From 1990 to the Long Term,” in Into the EU, Policy Frameworks in Central Europe, ed. by R. A. Feldman and C. M. Watson (Washington: International Monetary Fund), pp. 1333.

    • Search Google Scholar
    • Export Citation
  • Drummond, P., 2002, “Inflation Targeting in Hungary: Implementation and Policy Changes,IMF Selected Issues Paper SM/02/135 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Duffy, J., and C. Papageorgiou, 2000, “A Cross-Country Empirical Investigation of the Aggregate Production Function Specification,Journal of Economic Growth, Vol. 5, pp. 87120.

    • Search Google Scholar
    • Export Citation
  • Égert, B., 2002a, “Estimating the Impact of the Balassa-Samuelson Effect on Inflation and the Real Exchange Rate During the Transition,Economic Systems, Vol. 26, No. 1, pp. 116.

    • Search Google Scholar
    • Export Citation
  • Égert, B., 2002b, “Investigating the Balassa-Samuelson Hypothesis in the Transition. Do We Understand What We See?”, A Panel Study, Economics of Transition, Vol. 10, No. 2, pp. 273309.

    • Search Google Scholar
    • Export Citation
  • Garibaldi, P., and others, 2002, “What Moves Capital to Transition Countries?”, IMF Working Paper 02/64 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Ghei, N., and L. Pritchett, 1999, “The Three Pessimisms: Real Exchange Tate and Trade Flows in Developing Countries,” in Exchange Rate Misalignment, Concepts and Measurement for Developing Countries, ed. by P. Montiel and L. Hinkle (Washington: The World Bank), pp. 46796.

    • Search Google Scholar
    • Export Citation
  • Halpern, L., and C. Wyplosz, 1997, “Equilibrium Exchange Rates in Transition Economies,Staff Papers, International Monetary Fund, Vol. 4, No. 4, pp. 43061.

    • Search Google Scholar
    • Export Citation
  • Halpern, L., 2001, “Economic Transformation and Real Exchange Rates in the 2000s: The Balassa-Samuelson Connection,Mimeo, University of Geneva.

    • Search Google Scholar
    • Export Citation
  • Hausmann, R., and E. Fernández-Arias, 2000, “Foreign Direct Investment: Good Cholesterol?”, Working Paper No. 417, Inter-American Development Bank, Research Department.

    • Search Google Scholar
    • Export Citation
  • Hinkle, L., and P. Montiel, 1999, “Exchange Rate Misalignment: An Overview,in Exchange Rate Misalignment, Concepts and Measurement for Developing Countries, ed. by L. Hinkle and P. Montiel (New York: Oxford University Press), pp. 137.

    • Search Google Scholar
    • Export Citation
  • Hviding, K., 1999, Potential Output Growth in Hungary, IMF Staff Country Report No. 99/27 (Washington: International Monetary Fund).

  • Isard, P., and H. Faruqee, 1998, “Exchange Rate Assessment, Extensions of the Macroeconomic Balance Approach” IMF Occasional Paper No. 167 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Isard, P., and others, 2001, “Methodology for Current Account and Exchange Rate Assessment” IMF Occasional Paper No. 209 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Jakab, Z., and M. Kovács, 1999, “Determinants of Real-Exchange Rate Fluctuations in Hungary,MNB Working Paper, No. 1999/6.

  • Jakab, Z., 2002, “Hungary in the NIGEM Model,National Bank of Hungary Working Paper, No. 2002/3.

  • Jakab, Z., M. Kovács, and S. Lörincz, 2000, “Forecasting Hungarian Export Volume,National Bank of Hungary Working Paper, No. 2000/4.

    • Search Google Scholar
    • Export Citation
  • Kovács, M. (ed.), 2002, “On the Estimated Size of the Balassa-Samuel son Effect in Five Central and Eastern European Countries,National Bank of Hungary Working Paper, No. 2002/5.

    • Search Google Scholar
    • Export Citation
  • Lipschitz, L., T. Lane, and A. Mourmouras, 2002, “Capital Flows to Transition Economies: Master or Servant?”, IMF Working Paper 02/01 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Magya Nemzeti Bank (MNB), 2002a, Report on Financial Stability, December.

  • Magya Nemzeti Bank (MNB), 2002b, Background Information on Methodological Changes in the BOP Statistics, November.

  • Marsh, I., and P. Tokarick, 1994, “Competitiveness Indicators: A Theoretical and Empirical Assessment,IMF Working Papers 94/29 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Montiel, P., and L. Hinkle, 1999, “Exchange Rate Misalignment: An Overview, in Exchange Rate Misalignment, Concepts and Measurement for Developing Countries, ed. by P. Montiel and L. Hinkle (Washington: The World Bank), pp. 137.

    • Search Google Scholar
    • Export Citation
  • OECD, 2000, OECD Economic Surveys: Hungary, OECD: Paris.

  • Pelkmans, J., D. Gros, and J. Ferrer, 2000, “Long-Run Economic Aspects of the European Union’s Eastern Enlargement,WRR Scientific Council for Government Policy Working Paper, No. 109.

    • Search Google Scholar
    • Export Citation
  • Reinhart, C., 1995, “Devaluation, Relative Prices, and International Trade,Staff Papers, International Monetary Fund, Vol. 42, No. 2, pp. 290312.

    • Search Google Scholar
    • Export Citation
  • Riboud, M., C. Sanchez-Páramo, and C. Silva-Jauregui, 2002, “Does Eurosclerosis Matter? Institutional Reform and Labor Market Performance in Central and Eastern European Countries in the 1990s,Social Protection Discussion Paper, No. 0202 (Washington: The World Bank).

    • Search Google Scholar
    • Export Citation
  • Rother, C., 2000, The Impact of Productivity Differentials on Inflation and the Real Exchange Rate: An Estimation of the Balassa-Samuelson Effect in Slovenia, IMF Staff Country Report, No. 00/56, (Washington: International Monetary Fund) pp. 2639.

    • Search Google Scholar
    • Export Citation
  • Samuelson, P., 1964, “Theoretical Notes on Trade Inflation,Review of Economics and Statistics Vol. 46, pp. 14554.

  • Senhadij, A., 1997, “Time-Series Estimation of Structural Import Demand Equations: A Cross-Country Analysis,IMF Working Paper 97/132 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Sinn, H.-W. and M. Reutter, 2001, “The Minimum Inflation for Euroland, NBER Working Paper, No. 8085.

  • United Nations, 2002, “Transnational Corporations an Export Competitiveness,” World Investment Report, (New York: United Nations)

  • Vocke, M., 2001, “A Note on External Competitiveness,IMF Selected Issues Paper SM/01/121 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Wagner, N., 2002, The Demographic Shock and Hungary’s Pension System, IMF Staff Country Report, 02/109 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Whelan, K., 2000, “A Guide to the Use of Chain Aggregated NIPA Data,Mimeo, Federal Reserve Board.

1

Prepared by Helge Berger.

2

In addition to Hungary, the CECs are the Czech Republic, Poland, the Slovak Republic, and Slovenia.

3

It should be noted that the averages displayed in Figures 1 and 2 mask important within-group heterogeneity. This pertains both to the euro area and the CEC4 category. (The data on inflation in Figure 2, which deliberately excludes three-digit observations for Slovenia before 1993, is a particular stark example.) As a rule, however, the stylized facts emerging on the relative pattern of Hungary’s economic performance are well captured by the exercise.

4

Note that while the general statements hold, the year-to-year pattern of FDI may well look different, with regard to the fall off in recent years, if the series were adjusted to exclude privatization (which, as mentioned above, mainly occurred early on) and include reinvested earnings (though data are unavailable).

5

The average differential between CPI inflation in Hungary and the weighted CEC4 figure is 3.3 percentage points during 1990-2002.

6

This is not the result of averaging: Hungary’s current account deficits were by far the largest within the CEC group between 1993 and 1995.

7

The recent switch from settlement-based to custom-based data in Hungary complicates the comparison of current account data across time. Figure 2 reports revised data starting in 1995. For a detailed description of the changes, see the press release of February 25, 2003 and the note on methodological changes referenced therein, of the Magyar Nemzeti Bank (the National Bank or MNB).

8

The authorities have announced their intention to include these earnings in 2004. Doing so would imply an outflow of income (increasing the current account deficit) accompanied by a simultaneous inflow in FD1 (which improves the financial account balance).

9

Hungary has tended to have higher outward FDI flows than the other CECs. Since outward FDI can be considered a positive development, Figure 2 (Panel D) reports gross rather than net FDI inflows to avoid an inappropriate downward bias in the picture.

10

Net FDI flows in the case of Hungary amounted to less than 1 percent of GDP in 2002. The weighted average for the CEC4 was about 4¾ percent.

11

An upward movement in Figure 3 implies a real appreciation.

12

Using a VAR approach, Dibooglu and Kutan (2000) find that Hungary’s real exchange rate—as in many industrial countries—mainly reflects real shocks.

13

In a recent assessment of the evidence for Hungary, the MNB concludes that the effect should be in the range between 1 and 2 percent (Kovács 2002). Égert (2002a/b) argues along similar lines. Other contributions to the debate include Halpern and Wyplosz (1997, 2001), Jakab and Kovács (1999), Pelkmans and others (2000), Rother (2000), Sinn and Reuter (2001), Coricelli and Jazbec (2001), De Broeck and Sløk (2001), and Backé and others (2002).

14

Vocke (2001) finds ULC-based measures, while not without drawbacks, one of the most appropriate for use as a competitiveness measure for Hungary.

15

ULC-based REER data from the MNB is only available starting in 1995. All data is quarterly.

17

As Vocke (2001) observes, between 1996 and 2000 Hungary’s overall ranking improved more than any other country included the CEC. In the 2002/03 rankings, Slovenia is the only transition economy that consistently reaches (marginally) better rankings than Hungary.

18

The available data for 2002 points to a difference of about 1/15 percentage point in market share—not a trivial amount in economic terms: were the Hungarian market share to decline to the level of the CEC4 average, Hungarian exports would decrease by about 7.5 percent of GDP.

19

See Daseking (2001) for a similar exercise for Poland. Duffy and Papageorgiou (2000) provide a critical empirical assessment of the—analytically convenient—Cobb-Douglas framework that is used.

20

The estimated a relationship is  =(7. GAP. S +14. GAP)/1000, where  represents TFP growth, S average school years, and GAP the Hungarian-U.S. income gap. For the current exercise, S is set at about 9.7 and GAP is estimated at 2.9 on a 2001 PPP basis. The former figure takes into account an anticipated increase in the longer term suggested by World Bank forecasts.

21

Excluding consumer durables, which amount to about 0.3 percent of GDP.

22

Data for 1997—see Whelan (2000).

23

The OECD (2000) follows a similar approach.

24

This assumption is broadly in line with the average share of residential construction in investment flows. The share of residential construction in the United States capital stock is about 40 percent.

25

See Daseking(2001) for a related discussion for Poland.

27

The simulation below is based on figures provided by the authorities. The projections have been adjusted to reflect a somewhat more conservative view on the development of the participation rate in the medium term in line with World Bank estimates of Hungarian labor supply. The adjustment reduces the average employment growth by about 0.75 percentage points over the 2002-10 period compared to the original projection provided by the Ministry of Finance.

28

Hungary’s aging problem is somewhat less severe than in other EU and transition economies, not least because of the relatively shorter life expectancy of the Hungarian population (see Wagner 2002).

29

Conceptually, a change in the constant capital-GDP ratio underlying the simulation is equivalent to changing the depreciation rate. For instance, reducing the capital-GDP ratio from 2.4 to 2.05 produces a time path for the current account similar to that implied by the original ratio and a depreciation rate of 6 percent, and increasing the ratio to 2.75 yields a curve closely resembling the one with a depreciation rate of 9 percent. While this leaves the possibility of two “wrong” assumptions canceling each other out, there is also a chance for cumulating mistakes. Therefore, the error band around the results of the baseline scenario could be rather wide.

30

Hausmann and Fernandez-Aria (2000) argue that other forms for capital inflows share many of the positive attributes of foreign equity inflows. Instead equity inflows might simply reflect higher risk and poorly functioning markets for domestic financing. This hardly pertains to the state of Hungary’s credit and equity markets, however.

31

See Beaumont (1999) for a more elaborate discussion. Jakab and Kovàcs (2002) note that the transfer of know-how could also come through other sources, mitigating some of the sustainability issues potentially associated with the observed decline in FDI inflows.

32

Moody’s commented on its decision that the upgrades reflected its view that the process of economic and financial integration of these countries with the EU was virtually irreversible.

33

Reacting to rising government deficits, Standard & Poor recently downgraded Hungary, however.

34

During 1996-2002, equity FDI investments averaged 3.1 percent of GDP, while average non-equity FDI was 1.0 percent and portfolio equity investment 0.8 percent of GDP. Note that the data underlying Figure 10 excludes reinvested earnings.

35

See, among others, the recent analysis of FDI inflows by Garibaldi and others (2002) and Arvanitis (2002). Carlson and Hernandez (2002) discuss the determinants of the structure of overall capital inflows.

36

The average rank of the CEC4 is 40.5.

37

See Beaumont (1999) for an earlier analysis along similar lines.

39

The specific model used to calculate this result is not presented here, but is similar to the results reported in Chinn and Prasad (2000, Table A, column (2)). Estimates of the model were provided by the CGER group and are available on request.

40

See Isard and Faruqee (1998) and Isard and others (2001) for a detailed discussion of the CGER approach. CGER is short for the IMF’s Coordinating Group on Exchange Rate Issues.

41

In addition, the approach does not allow for an explicit feedback from the real exchange rate to the variables determining the benchmark current account.

42

See Senhadij (1997) for a different view. The modifications of the CGER model described by Isard and others (2001, Appendix II) assume that 80 percent of the impact of a REER shock takes place within the first three years.

44

For instance, Ghei and Pritchett (1999), based on an extensive survey of the existing literature, place the long-run price elasticity of imports at -0.9 for industrial and at -0.7 for developing countries. Reinhart (1995), too, finds that “traditional” factors have a significant impact on trade-flows.

45

In the case of Hungary, using 1992-1999 data, Jakab and others (2000) find an elasticity of just -0.21 for ULC-based effective exchange rates. Beaumont (1999) estimates a trade-balance elasticity of 0.18 for the 1992-98 period. Vocke (2001) argues that these estimates could still be biased upwards due to the growing role of multinationals in Hungary’s external trade during the 1990s.

46

As mentioned earlier, example for such factors would be a temporary decline in tourism revenues or a transitory increase in oil prices. The underlying current account would be biased upward, if the starting point of the exercise were not adjusted accordingly.

47

The smallest current account deficit observed during the 1995-2002 period was 1.4 percent of GDP in 1997.

49

A benchmark in that vicinity would be in line with the views expressed in the MNB’s Report on Financial Stability (MNB 2002a). The report considers a current account deficit of about 5 percent of GDP excluding reinvested earnings as sustainablewhich translates roughly into a deficit of about 7.5 percent including reinvested earnings.

50

Sample selection was determined by data availability. Experiments with an extended data set including available data in 2001 yielded broadly similar results.

51

The net-impact of the squared and non-squared stock variable is strictly positive across the sample.

52

The regression controls for differences in inflation levels.

53

In part, this is due to the fact that the U.N. measure of potential already takes into account export performance.

54

Average (AVG) and standard deviation (STD) based on model (4) in the main text; based on 96 observations.

  • Alesina, A., and others, 2002, “Fiscal Policy, Profits and Investment,American Economic Review (June) pp. 58189.

  • Alesina, A., 1998, “Tales of Fiscal Adjustment,” Economic Policy: A European Forum (United Kingdom).

  • Alesina,A., and R. Perotti, 1997, “The Welfare State and Competitiveness,American Economic Review, Vol. 87, No. 5 (December) pp. 92139.

    • Search Google Scholar
    • Export Citation
  • Alesina,A., 1995a, “The Political Economy of Budget Deficits,IMF Staff Papers, International Monetary Fund, Vol. 42.

  • Alesina,A., 1995b, “Fiscal Expansions and Adjustments in OECD Countries,Economic Policy, No. 21 (August).

  • Bradley, J., and K. Whelan, 1997, “The Irish Expansionary Fiscal Contraction,Economic Modeling, Vol. 14.

  • Brumby, J., and M. Cangiano, 2001, “Public Expenditure Management Reform and Fiscal Consolidation in OECD Countries,paper presented at the Fifth International Conference Institutions in Transition in Otocec, Slovenia, July.

    • Search Google Scholar
    • Export Citation
  • Buti, M., and A. Sapir, ed., 1998, Economic Policy in EMU: A Study by the European Commission Services (New York: Oxford University Press).

    • Search Google Scholar
    • Export Citation
  • Buti, M., and others, 1997, “Budgetary Policies During Recessions: Retrospective Application of the “Stability and Growth Pact” to the Post-war Period,” Economic Papers No. 121 (Brussels: European Commission, Directorate-General for Economic and Financial Affairs).

    • Search Google Scholar
    • Export Citation
  • Diamond, J. 2003, “Performance Budgeting: Managing the Reform Process,IMF Working Paper 03/33 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Feldman, R., and M. Watson, 2002, Into the EU: Policy Frameworks in Central Europe (Washington: International Monetary Fund).

  • Hagemann R., and N. Koliadina, 2000, Fiscal Policy in Ireland: Trends and Prospects IMF Staff Country Report No. 00/99 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Hemming, R., and M. Kell, 2001, “Promoting Fiscal Responsibility: Transparency, Rules, and Independent Fiscal Authorities,in Fiscal Rules, Banca d’ltalia Research Department Public Finance Workshop, Perugia, Italy, February.

    • Search Google Scholar
    • Export Citation
  • Hemming R., M. Kell, and S. Mahfouz, 2002, “The Effectiveness of Fiscal Policy in Stimulating Economic Activity-A Review of the Literature.IMF Working Paper No. 02/208 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Honohan, P., 1999, “Fiscal Adjustment and Disinflation,” in Understanding Ireland’s Economic Growth, ed. by Frank Barry, (London: Macmillan).

    • Search Google Scholar
    • Export Citation
  • Hungary, Ministry of Finance, 2002, “Pre-Accession Economic Programme of Hungary-2002” (Budapest).

  • Kell, M., 2001, “An Assessment of Fiscal Rules in the U.K.,IMF Working Paper No. 01/91 (Washington: International Monetary Fund).

  • Institute for Fiscal Studies, 1998, Options for 1997: The Green Budget (London).

  • Organization for Economic Cooperation and Development, 1987, Economic Surveys: Ireland (Paris).

  • Organization for Economic Cooperation and Development, 1998, Economic Surveys: Sweden (Paris).

  • Organization for Economic Cooperation and Development, 1995, Budgeting for Results (Paris).

  • Organization for Economic Cooperation and Development, 1996, Managing Structural Deficit Reduction—A Synthesis of Member Countries Experiences (Paris).

    • Search Google Scholar
    • Export Citation
  • Organization for Economic Cooperation and Development, 2002a, “Overview of Results Focused Management and Budgeting in OECD Member Countries,” at the Expert Meeting on the Quality of Public Expenditure: Implementation Challenges in Result Focused Budgeting and Management, Paris, February.

    • Search Google Scholar
    • Export Citation
  • Organization for Economic Cooperation and Development, 2002b, “Governing for Results,” at the 25th Session of the Public Management Committee, Chateau de la Muette, Paris, March.

    • Search Google Scholar
    • Export Citation
  • McAleese, D., 1989, “Adjustment and External Shocks in Ireland,Working Paper Series 262, International Economic Department (Washington: The World Bank).

    • Search Google Scholar
    • Export Citation
  • Perotti R. 1996a, “Fiscal Consolidation in Europe: Composition Matters,AEA Papers and Proceedings, May.

  • Perotti R., 1998, “The Political Economy of Fiscal Consolidations,Scandinavian Journal of Economics Vol. 100, No.1, pp. 36795.

  • Purfield, C., 2003, “Fiscal Adjustment in Transition: Evidence from the 1990s,IMF Working Paper No. 03/36 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Roseveare, D., 2002, “Enhancing the Effectiveness of Public Expenditure in Sweden,Economics Department Working Papers No.345 (Paris: Organization for Economic Cooperation and Development).

    • Search Google Scholar
    • Export Citation
  • Van den Noord, P., 2002, “Managing Public Expenditure: the U.K. Approach,Economics Department Working Papers No. 341 (Paris: OECD).

    • Search Google Scholar
    • Export Citation
  • Von Hagen, J., and others, 2001, “Budgetary Consolidation in EMU,CEPR Discussion Paper No. 48 (London: Centre for Economic Policy Research).

    • Search Google Scholar
    • Export Citation
1

Prepared by Paulo Drummond, Eva Jenkner, and Gerd Schwartz.

2

See Government of the Republic of Hungary, “Pre-Accession Economic Programme of Hungary—2002,” Budapest, August 2002.

3

Structural (cyclically adjusted) primary balances also improved during these consolidation periods—by 5 percent on average—with an annual adjustment of 1 percent of GDP. Some countries achieved much stronger adjustments, including Sweden, Italy, and Greece.

4

As explained in SM/03/116, the fiscal adjustment in 2003 relies mostly on one-off factors (3¾ percentage points of GDP) and lower investment spending.

5

Defined as the cyclically adjusted primary balance.

6

Belgium, Austria and Portugal. Ireland is included in the group of countries with revenue-based adjustment in the early 90s, but it implemented the bulk of its adjustment in the late 80s. See a separate discussion on the Irish adjustment path in Box 2.

7

Germany, Greece, Spain, France, Italy, and the Netherlands.

8

These changes may have led to an understatement of wage spending in the budget. For example, “purchases” of health care services, which are classified as expenditures on goods and services, in large part cover wage costs.

9

Health spending as a percentage of GDP recovered slightly during the 1990s, but per capita spending remains below the European average.

10

Von Hagen and others, (2001) define fiscal consolidations as episodes in which either the cyclically-adjusted government balance increased (i.e., a smaller deficit or a larger surplus) by at least 1.25 percent of cyclically-adjusted GDP in two consecutive years, or the change in the cyclically-adjusted budget balance exceeded 1.5 of cyclically adjusted GDP in one year and was positive but less than 1.25 percent in the preceding year and in the subsequent year. A consolidation effort is deemed successful if, two years after the initial adjustment, the government budget balance stands at no less than 75 percent of the balance in the first year of the consolidation episode.

11

This finding is consistent with Purfield (2003), who reviews episodes of fiscal consolidation in 25 transition countries during 1992–2000 and finds evidence that policies which focused on expenditure reduction were more successful in addressing fiscal imbalances than those which relied on revenue increases.

12

For example, see von Hagen (2001), Alesina (1998), Alesina and Perotti (1995a and b).

14

As vocal and politically influential groups can be expected to resist expenditure cuts ex post, across-the-board cuts may not be credible ex ante.

16

The main excluded items were social security benefits, local government expenditures financed by its own revenues, and interest payments.

18

See Diamond (2003) for a more elaborate discussion.

19

The difference between outputs and outcomes is often a source of confusion. Outputs are defined as the goods and services provided by the government. Outcomes are the policy impacts on the community as a result of government outputs. For example, an output could bt the number of patients treated, whereas the outcome would be good public health.

20

Specifically, biennial spending reviews set out three-year plans for discretionary expenditure by departments, or Departmental Expenditure Limits (DELs). Departments have to bid for funds and have to enter so-called Public Service Agreements (PSAs), which hold them accountable for achieving targets set out in the contract. Technical notes set out in detail how each target will be measured, specifying data sources, the baseline, and any potential ambiguities. Progress against targets is presented quarterly to the Cabinet committee, and published annually. Resources in the following budget round are allocated in light of past performance.