Abstract
This note compares patterns of domestic investment in Serbia with those in other central and eastern European countries, noting the relationships with external balances. The structure of participation and employment rates suggests a need for analysis of the impact of labor market institutions on youth and women. A further focus on redeployment services would be appropriate. The Serbian banking system, the implications of the structure of Serbia’s economy, the operational framework of monetary policy, and the adoption of an inflation targeting regime have been discussed.
V. Economic Structure and Choice of Exchange Rate Regime16
A. Introduction
1. This note considers the implications of the structure of Serbia’s economy for its choice of exchange rate regime. The discussion is motivated by the authorities’ intention to accommodate greater volatility in the exchange rate. It concludes that though a number of the characteristics of the Serbian economy point towards the desirability of a relatively stable nominal exchange rate, others point towards the need for flexibility. It therefore suggests that a flexible regime, managed with little or no discretionary foreign exchange intervention, but with other policies aimed to stabilize the nominal exchange rate to a significant degree could provide the best regime for Serbia.
B. Structural Factors
2. Structural factors which influence the choice of exchange rate regime include:
Openness: The more open a country, and the larger and synchronized its trade is with its main trading partner, the more it would gain from a stable (e.g., fixed) exchange rate with that partner.
Diversification: Diversified economies are less exposed to terms-of-trade shocks and can thus more easily live with a fixed exchange rate.
Capital flows: The more an economy is integrated into international capital markets, the greater its potential need for a flexible exchange rate to cushion against shocks to capital flows. This includes scope for residents to acquire foreign currency denominated assets, either in the domestic banking system, or, if the exchange control regime allows, externally.
Shocks: If shocks are predominantly real, flexible exchange rates are more appropriate as they provide a useful adjustment mechanism. If shocks are predominantly nominal, a credible and stable fixed exchange rate regime may usefully reduce the incidence of such shocks. Countries with histories of high or hyperinflation may come into the latter category.
Dollarization: If high, this may reflect agents’ judgment that inflation is more volatile than the real exchange rate, and is thus symptomatic of the high incidence of nominal over real shocks. These characteristics, as noted above, tend to call for greater stability of the nominal exchange rate.
C. Serbia’s Structural Characteristics in International Context
3. Serbia’s characteristics are compared with a large number of countries in Table 1.17 Countries are classified and ranked in regard to the structural criteria noted above amongst others, and Serbia’s characteristics are reported alongside. Thus, on trade openness, the median for countries’ shares of exports and imports in GDP is 55, while it is 66 for Serbia. This relatively high percentage for Serbia is reflected in the far right hand column as suggesting that Serbia should incline to more stability in its exchange rate. The same exercise is repeated for the other criteria.
Serbia: Quantitative Assessment of the Choice of the Exchange Rate Regime
52 developed, emerging market, and developing countries in Husain (2004).
1 = case for peg; 5 = case for float.
Serbia: Quantitative Assessment of the Choice of the Exchange Rate Regime
| All Countries 1/ | Serbia | |||||||
|---|---|---|---|---|---|---|---|---|
| Time period | Range | Median | Time period | Score | Quantile 2/ | |||
| Economic Integration | ||||||||
| Trade orientation (exports and imports of goods and services in percent of GDP) | 1999-2002 | 18-272 | 55 | 1999-2002 | 66 | 2 | ||
| Trade pattern concentration (export share of key partner currency in percent of total exports) | 1989-91 | 23-77 | 38 | 1999-2001 | 54 | 2 | ||
| Cyclical synchronicity with trade partner (real GDP growth correlation with key export partner) | 1993-2003 | 0.1-1.0 | 0.2 | 1993-2003 | −0.25 | 5 | ||
| 2000-05 | 0.38 | 2 | ||||||
| Financial integration | ||||||||
| Inclusion in major indices (MSCI, EMBI+, S&P/IFCG Composite, S&P/IFCG Frontier Composite) | 2005 | 1 | 3 | |||||
| Stock market capitalization (in percent of GDP) | end-2003 | 0.5-455 | 29 | Q3 2005 | 32 | 3 | ||
| Stock market annual turnover (in percent of GDP) | end-2003 | 0.1-198 | 1 | Q3 2005 | 2 | 4 | ||
| Financial development (ratio of broad money to GDP) | 1992-2002 | 9-227 | 46 | 1997-2002 | 17 | 1 | ||
| 1997-2004 | 18 | 1 | ||||||
| 2004 | 24 | 2 | ||||||
| Capital flows, net (ratio to GDP) | 1995-2003 | 0-2 | 0.1 | 1997-2003 | 6 | 5 | ||
| 2001-05 | 13 | 5 | ||||||
| Economic diversification | ||||||||
| Terms of trade volatility (coefficient of variation) | 1998-2002 | 0.01-0.18 | 0.1 | |||||
| Primary commodities exports (share in total exports, in percent) | 1990-99 | 2-98 | 21 | 2004-05 | 41 | 4 | ||
| Primary commodities exports (ratio to GDP, in percent) | 1990-99 | 1-40 | 5 | 2004-05 | 7 | 4 | ||
| Monetary and real shocks | ||||||||
| Trade versus capital flows (ratio of cumulative gross capital flows to trade flows, in percent) | 1998-2003 | 6-154 | 18 | 1998-2003 | 19 | 3 | ||
| 2001-05 | 19 | 3 | ||||||
| Monetary volatility (coefficient of variation of broad money-to-GDP ratio) | 1998-2002 | 0.02-0.26 | 0.06 | 1998-2002 | 0.19 | 1 | ||
| 2001-05 | 0.20 | 1 | ||||||
| Monetary versus Real shocks (ratio of coefficient of variation of broad money-to-GDP to that of real GDP) | 1993-2002 | 0.0-1.2 | 0.13 | 2001-05 | 0.46 | 2 | ||
| Credibility | ||||||||
| Inflation history-8 percent (proportion of months in which y-o-y inflation exceeded 8 percent, in percent) | 1994-2003 | 0-100 | 29 | 1995-2003 | 98 | 1 | ||
| 2001-05 | 90 | 1 | ||||||
| Inflation history-10 percent (proportion of months in which y-o-y inflation exceeded 10 percent, in percent) | 1994-2003 | 0-100 | 15 | 1995-2003 | 94 | 1 | ||
| 2001-05 | 85 | 1 | ||||||
| Fear of Floating-type effects | ||||||||
| Dollarization (Reinhart, Rogoff, and Savastano 2003 index of dollarization) | 1996-2001 | 0-22 | 7 | 2001-05 | 20 | 1 | ||
| Balance sheet effects (correlation of percent changes of real GDP and NEER) | 1993-2003 | −0.9 - +0.9 | 0 | 1997-2004 | 0.1 | 3 | ||
| Exchange rate pass-through (correlation of y-o-y percent changes of NEER and CPI) (direct) | 1998-2003 | −0.8 - +0.8 | −0.1 | 1998-2003 | −0.6 | 2 | ||
| 2001-05 | −0.8 | 1 | ||||||
| (lagged) | 1998-2003 | −0.9 - +0.8 | −0.1 | 1998-2003 | −0.8 | 1 | ||
| 2001-05 | −0.9 | 1 | ||||||
52 developed, emerging market, and developing countries in Husain (2004).
1 = case for peg; 5 = case for float.
Serbia: Quantitative Assessment of the Choice of the Exchange Rate Regime
| All Countries 1/ | Serbia | |||||||
|---|---|---|---|---|---|---|---|---|
| Time period | Range | Median | Time period | Score | Quantile 2/ | |||
| Economic Integration | ||||||||
| Trade orientation (exports and imports of goods and services in percent of GDP) | 1999-2002 | 18-272 | 55 | 1999-2002 | 66 | 2 | ||
| Trade pattern concentration (export share of key partner currency in percent of total exports) | 1989-91 | 23-77 | 38 | 1999-2001 | 54 | 2 | ||
| Cyclical synchronicity with trade partner (real GDP growth correlation with key export partner) | 1993-2003 | 0.1-1.0 | 0.2 | 1993-2003 | −0.25 | 5 | ||
| 2000-05 | 0.38 | 2 | ||||||
| Financial integration | ||||||||
| Inclusion in major indices (MSCI, EMBI+, S&P/IFCG Composite, S&P/IFCG Frontier Composite) | 2005 | 1 | 3 | |||||
| Stock market capitalization (in percent of GDP) | end-2003 | 0.5-455 | 29 | Q3 2005 | 32 | 3 | ||
| Stock market annual turnover (in percent of GDP) | end-2003 | 0.1-198 | 1 | Q3 2005 | 2 | 4 | ||
| Financial development (ratio of broad money to GDP) | 1992-2002 | 9-227 | 46 | 1997-2002 | 17 | 1 | ||
| 1997-2004 | 18 | 1 | ||||||
| 2004 | 24 | 2 | ||||||
| Capital flows, net (ratio to GDP) | 1995-2003 | 0-2 | 0.1 | 1997-2003 | 6 | 5 | ||
| 2001-05 | 13 | 5 | ||||||
| Economic diversification | ||||||||
| Terms of trade volatility (coefficient of variation) | 1998-2002 | 0.01-0.18 | 0.1 | |||||
| Primary commodities exports (share in total exports, in percent) | 1990-99 | 2-98 | 21 | 2004-05 | 41 | 4 | ||
| Primary commodities exports (ratio to GDP, in percent) | 1990-99 | 1-40 | 5 | 2004-05 | 7 | 4 | ||
| Monetary and real shocks | ||||||||
| Trade versus capital flows (ratio of cumulative gross capital flows to trade flows, in percent) | 1998-2003 | 6-154 | 18 | 1998-2003 | 19 | 3 | ||
| 2001-05 | 19 | 3 | ||||||
| Monetary volatility (coefficient of variation of broad money-to-GDP ratio) | 1998-2002 | 0.02-0.26 | 0.06 | 1998-2002 | 0.19 | 1 | ||
| 2001-05 | 0.20 | 1 | ||||||
| Monetary versus Real shocks (ratio of coefficient of variation of broad money-to-GDP to that of real GDP) | 1993-2002 | 0.0-1.2 | 0.13 | 2001-05 | 0.46 | 2 | ||
| Credibility | ||||||||
| Inflation history-8 percent (proportion of months in which y-o-y inflation exceeded 8 percent, in percent) | 1994-2003 | 0-100 | 29 | 1995-2003 | 98 | 1 | ||
| 2001-05 | 90 | 1 | ||||||
| Inflation history-10 percent (proportion of months in which y-o-y inflation exceeded 10 percent, in percent) | 1994-2003 | 0-100 | 15 | 1995-2003 | 94 | 1 | ||
| 2001-05 | 85 | 1 | ||||||
| Fear of Floating-type effects | ||||||||
| Dollarization (Reinhart, Rogoff, and Savastano 2003 index of dollarization) | 1996-2001 | 0-22 | 7 | 2001-05 | 20 | 1 | ||
| Balance sheet effects (correlation of percent changes of real GDP and NEER) | 1993-2003 | −0.9 - +0.9 | 0 | 1997-2004 | 0.1 | 3 | ||
| Exchange rate pass-through (correlation of y-o-y percent changes of NEER and CPI) (direct) | 1998-2003 | −0.8 - +0.8 | −0.1 | 1998-2003 | −0.6 | 2 | ||
| 2001-05 | −0.8 | 1 | ||||||
| (lagged) | 1998-2003 | −0.9 - +0.8 | −0.1 | 1998-2003 | −0.8 | 1 | ||
| 2001-05 | −0.9 | 1 | ||||||
52 developed, emerging market, and developing countries in Husain (2004).
1 = case for peg; 5 = case for float.
4. Drawing on Table 1, these indicators for Serbia are summarized on Table 2. Though a number of indicators point towards flexibility, the scoring indicates that an unweighted examination of criteria tends to point towards the desirability of a stable exchange rate. This information is supplemented in Table 3 with more detailed data reported for Serbia. In summary, the high euroization, the high exchange rate pass-through, the long history of inflation, the high monetary volatility, and the limited financial intermediation place Serbia as a candidate for a stable exchange rate, relative to the other countries in the sample. Moreover, its relatively high trade openness, concentrated trade pattern (with respect to the euro area or countries pegged to the euro), synchronized cycle with trade partners, and the prevalence of monetary over real shocks also constitute factors in favor of stability. On the other hand, the presence of large capital flows argues in favor of flexibility, as well as the relatively high share of primary commodities exports and the high stock market turnover (which reflects large capital flows).
Serbia: Ranking of Arguments
Relative to 52 other countries in Husain (2004).
Serbia: Ranking of Arguments
| Indicator | Quantile 1/ |
|---|---|
| Case for a peg | |
| High dollarization | 1 |
| High exchange rate pass-through | 1 |
| History of inflation | 1 |
| High monetary volatility | 1 |
| Limited financial development | 1 |
| High trade orientation | 2 |
| Concentrated trade pattern | 2 |
| Synchronized cycle with major trade partners | 2 |
| Prevalence of monetary shocks | 2 |
| Case for a float | |
| Large capital flows | 5 |
| High share of primary commodities exports | 4 |
| High stock market turnover | 4 |
| No strong case | |
| Average balance sheet effects on real activity | 3 |
| Inclusion in only one major emerging market index | 3 |
| Average trade versus capital flows | 3 |
| Average stock market capitalization | 3 |
Relative to 52 other countries in Husain (2004).
Serbia: Ranking of Arguments
| Indicator | Quantile 1/ |
|---|---|
| Case for a peg | |
| High dollarization | 1 |
| High exchange rate pass-through | 1 |
| History of inflation | 1 |
| High monetary volatility | 1 |
| Limited financial development | 1 |
| High trade orientation | 2 |
| Concentrated trade pattern | 2 |
| Synchronized cycle with major trade partners | 2 |
| Prevalence of monetary shocks | 2 |
| Case for a float | |
| Large capital flows | 5 |
| High share of primary commodities exports | 4 |
| High stock market turnover | 4 |
| No strong case | |
| Average balance sheet effects on real activity | 3 |
| Inclusion in only one major emerging market index | 3 |
| Average trade versus capital flows | 3 |
| Average stock market capitalization | 3 |
Relative to 52 other countries in Husain (2004).
Serbia: Exchange Rate Regime Considerations, 2005 1/
Serbia was benchmarked against 52 other countries, as in Husain, 2004, mimeo, and Eken and al., 2005, IMF Country Report 05/419. Most figures refer to 2005, or to 2001-05.
Serbia: Exchange Rate Regime Considerations, 2005 1/
| Indicator | Score | Units |
|---|---|---|
| Case for a peg | ||
| High dollarization | 70 | Share of fx-deposits in total |
| High exchange rate pass-through | 0.9 | On core inflation, one year |
| History of inflation | 85 | Percent of months with y-o-y inflation higher than 10 percent in last 5 years |
| High monetary volatility | 0.2 | Coefficient of variation of M2 to GDP |
| Limited financial development | 27 | M2 to GDP |
| High trade orientation | 76 | Exports and imports of G&S in percent of GDP |
| Concentrated trade pattern | 54 | Export share to euro-linked partner countries |
| Synchronized cycle with major trade partners | 0.4 | Real GDP growth correlation with euro area |
| Prevalence of monetary shocks | 0.5 | Ratio of coefficient of variation of M2-to-GDP to that of real GDP |
| Case for a float | ||
| Large capital flows | 18 | Net capital flows to GDP |
| High share of primary commodities exports | 41 | In percent of total exports |
| High stock market turnover | 2 | In percent of GDP |
| No strong case | ||
| Average balance sheet effects on real activity | 0.1 | Correlation of real GDP and NEER changes |
| Inclusion in only one major emerging market index | EMBI+ | |
| Average trade versus capital flows | 19 | Cumulative gross capital flows to trade flows, in percent |
| Average stock market capitalization | 32 | In percent of GDP |
Serbia was benchmarked against 52 other countries, as in Husain, 2004, mimeo, and Eken and al., 2005, IMF Country Report 05/419. Most figures refer to 2005, or to 2001-05.
Serbia: Exchange Rate Regime Considerations, 2005 1/
| Indicator | Score | Units |
|---|---|---|
| Case for a peg | ||
| High dollarization | 70 | Share of fx-deposits in total |
| High exchange rate pass-through | 0.9 | On core inflation, one year |
| History of inflation | 85 | Percent of months with y-o-y inflation higher than 10 percent in last 5 years |
| High monetary volatility | 0.2 | Coefficient of variation of M2 to GDP |
| Limited financial development | 27 | M2 to GDP |
| High trade orientation | 76 | Exports and imports of G&S in percent of GDP |
| Concentrated trade pattern | 54 | Export share to euro-linked partner countries |
| Synchronized cycle with major trade partners | 0.4 | Real GDP growth correlation with euro area |
| Prevalence of monetary shocks | 0.5 | Ratio of coefficient of variation of M2-to-GDP to that of real GDP |
| Case for a float | ||
| Large capital flows | 18 | Net capital flows to GDP |
| High share of primary commodities exports | 41 | In percent of total exports |
| High stock market turnover | 2 | In percent of GDP |
| No strong case | ||
| Average balance sheet effects on real activity | 0.1 | Correlation of real GDP and NEER changes |
| Inclusion in only one major emerging market index | EMBI+ | |
| Average trade versus capital flows | 19 | Cumulative gross capital flows to trade flows, in percent |
| Average stock market capitalization | 32 | In percent of GDP |
Serbia was benchmarked against 52 other countries, as in Husain, 2004, mimeo, and Eken and al., 2005, IMF Country Report 05/419. Most figures refer to 2005, or to 2001-05.
D. Discussion
5. Though useful, care is needed in interpretation of the results. For example, the evidence that Serbia exhibits monetary shocks is based on the volatility of velocity estimates compared to other countries. However, though velocity has changed significantly in Serbia since 2000, this is largely because of a sustained increase in depositor confidence in banks. Thus, though the change is reflected in this exercise as indicative of instability, the change in this case reflects a return to stability. However, the conclusion—that nominal shocks are high—may nevertheless be correct. In particular, an exercise like this requires somewhat arbitrary assumptions about the periods over which the indicators are measured. Here, the Serbian data reported are largely for the period since the late 1990s. But the most relevant period would be that which agents use to determine their own monetary behavior. That, of course, cannot be directly observed. But it may be inferred indirectly from Serbs’ choices regarding financial euroization. The persistence of the latter in the face of sharp reductions in inflation since 2000 largely reflects lingering skepticism about inflationary prospects due to the two hyperinflations in recent memory. Thus, arguably, the data sets used for assessments of the sort in the exercise above should be longer because behavior by agents in respect of financial euroization indicates that in their view, monetary instability is high, regardless of the recovery of confidence in banks reflected in the velocity of broad money since 2000.
6. A further difficulty concerns the reliability of parameter estimates. While, as noted above, the information contained in basic indicators reflects, in part, relatively arbitrary decisions about the time period over which the indicators are measured, parameter estimates—for factors such as pass through from exchange rates to inflation or the synchronization of activity with major trading partners—can be both period and methodology-specific. A variety of pass through estimates for Serbia are available and, as noted in Chapter 7, the variation in the estimates may reflect inadequacies with the techniques applied. With short data sets and multiple structural shocks both severely constraining the reliability of such parameter estimates, the implications for choice of exchange rate regime from estimates of behavioral parameters should be drawn with caution.
7. Structural change should also be taken into account. Basic indicators and parameter estimates, even when free of the concerns noted above, summarize information about the past. But structural change may be moving ahead rapidly, and in this context, the structural characteristics which will pertain—as opposed to those which pertained in the past—form the more pertinent basis on which to determine the choice of exchange rate regime. Of the various prospective structural changes, perhaps the key one for present purposes is increasing access to international financial markets. Access currently is limited by immediate political uncertainties and by the limited development of domestic financial markets. However, if those political difficulties are resolved, Serbia is likely to experience a rapid and significant deepening in its access to international capital markets. This factor would argue for progress towards greater flexibility in the exchange rate arrangements now so as to anticipate this development.
8. And the specific implications of structural characteristics—past or anticipated—for choice of exchange rate regime are not always straightforward. For example, the suggestion in the literature that the high past or prospective incidence of monetary shocks points towards a fix is qualified to the extent that other mechanisms might be available to reduce the incidence of such shocks. For example, while a fix is one “technology” available to policymakers to reduce the incidence of monetary shocks, the adoption of a strong “inflation targeting” regime could be another such technology, even though it typically anticipates a float. This distinction may be important, as in the case of Serbia, where a number of considerations—such as openness to capital flows—argue against a fix.
9. The exercise also requires some weighting of the various structural indicators. This issue is particularly acute when, as in the case of Serbia, the indicators point in different directions regarding the exchange rate regime. Given the large shocks to capital flows which emerging markets have experienced in recent years, echoed in market reassessments since Spring 2006, capital flows concerns should probably loom large in this relative weighting, pointing towards flexibility.
10. Finally, assessments such as those described may yield conclusions about the way in which a given exchange rate regime should be operated, as opposed to determining the choice of regime itself. Thus, as indicated above, Serbia’s structural characteristics point in various directions regarding the choice of exchange rate regime, some pointing more to a fix, and others to a float. But if, for example, the arguments for a float are considered to dominate other concerns, the other structural characteristics could nevertheless also be reflected in policies which are aimed to minimize the volatility in the exchange rate under that regime. In other words, the structural characteristics said to point towards a fix in fact point towards stability, whether that is achieved through foreign intervention—a fix—or by other means under a float. Those other means could include the strength of supportive fiscal policy—including the swiftness and high quality of its response to macroeconomic shocks—and policy towards the central monetary policy interest rate (See Chapter 6).
11. A float supported by fiscal and monetary policies targeting stability, may be the best fit for Serbia. With capital flows issues and political uncertainties looming large, the case for a float is strong. High euroization, openness, and other structural factors calling for stability should, however, be reflected in the way the regime is operated. In particular, strong and appropriately flexible supportive policies—abjuring discretionary intervention—alongside development of derivative markets to facilitate exchange risk hedging could aim to provide agents with a degree of stability in the external exchange rate appropriate to Serbia’s structural characteristics.
References
Eken, Sena, Abdourahmane Sarr, Jacques Bouhga-Hagbe, and Jérôme Vandenbussche, 2005, “Morocco: Exchange Rate Regime,” in Morocco—Selected Issues, IMF Country Report No. 05/419 (Washington: International Monetary Fund), pp. 41 –89.
Husain, Aasim M., 2004, “To Peg or Not to Peg—A Template for Assessing the Nobler,” (Washington: International Monetary Fund), unpublished.
Husain, Aasim M., 2006, “To Peg or Not to Peg—A Template for Assessing the Nobler,” IMF Working Paper WP/06/54 (Washington: International Monetary Fund).
Prepared by Eric Mottu.
See also Husain (2006). Whenever possible, we calculate the indicators for Serbia for a time frame as close as possible to that of the benchmark countries; we also report more recent figures for Serbia when those are significantly different from historical data.