Information about Europe Europa
Journal Issue
Share
Article

Ukraine: Selected Issues

Author(s):
International Monetary Fund
Published Date:
February 2007
Share
  • ShareShare
Information about Europe Europa
Show Summary Details

II. Monetary and Exchange rate Policy framework—where to go from here?

Core Questions, Issues, and Findings

  • How well is the hryvnia aligned with external fundamentals? In 2004, with the current-account surplus rising to some 10 percent of GDP, the hryvnia was viewed as significantly undervalued. This chapter’s reassessment—based on more recent data and using the macroeconomic balance approach—suggests that the hryvnia’s real effective undervaluation has narrowed sharply (¶61–66).
  • Where is Ukraine’s real exchange rate likely headed over the longer term? Results based on dollar-wage and purchasing-power-parity comparisons suggest that—as Ukraine’s per capita income catches up to levels in more advanced economies—its real effective exchange rate should appreciate substantially (¶71–76).
  • Why has Ukraine developed a strong affinity for its de facto peg? After the 1998 financial crisis, policy credibility was low, and the de facto peg served well as a monetary anchor. With a benign external environment and generally supporting incomes and fiscal policies, the peg became associated with a period of rapid real growth, while widespread dollarization underpins a fear-of-floating culture (¶81–84).
  • What would it take to make the peg work in Ukraine’s changing macroeconomic setting? The key requirements would be a fiscal policy oriented toward maintaining internal balance combined with a high degree of real wage and price flexibility to facilitate adjustment to shocks to external competitiveness (¶86–90).
  • What are the attractions of a more flexible exchange-rate regime in Ukraine’s changing macroeconomic setting? A more flexible exchange rate would help the NBU gain better control over inflation and facilitate adjustment to external shocks. Greater variability of the exchange rate should also discourage dollarization and excessive risk taking by unhedged borrowers (¶96–97).
  • Is the time auspicious for a move to a new monetary framework? Domestic and external conditions seem supportive: international reserve levels are comfortable, the exchange rate is better aligned with fundamentals, fiscal and incomes policies are set to tighten over the medium term, and conditions in international markets are benign. Moreover, the NBU has already advanced technical preparations toward introducing inflation targeting. However, the government needs to provide the NBU with a clearer mandate to pursue price stability as its main objective and spur domestic financial market development (¶98–99).

54. Ukraine’s monetary and exchange rate policy framework is at a crossroads. What constitutes an appropriate monetary and exchange rate policy framework for Ukraine has been the subject of a long-standing debate.24 Over the last few years, agreement has been reached in principle that Ukraine would benefit from shifting to a more flexible exchange rate and a monetary framework based on inflation targeting. But the appropriate pace for transition to a new framework remains controversial. At the same time, the debate has acquired new urgency given Ukraine’s rapidly changing external environment—see, for example, Chapter I on energy price shocks.

55. This chapter’s first section re-assesses Ukraine’s equilibrium exchange rate mainly based on the macroeconomic balance approach. Ukraine’s equilibrium exchange rate was last comprehensively assessed in the context of the 2004 Article IV consultation, but much has changed since. With the current-account surplus in 2004 having been projected to rise to some 10 percent of GDP, the analysis at that time pointed to a clear-cut case of substantial real undervaluation of the hryvnia. Since 2004, Ukraine’s external position has rebalanced substantially, and this section reviews recent external developments. The section’s re-assessment is mainly based on the IMF’s macroeconomic balance approach, but also cross-checked with purchasing-power-parity (PPP) and dollar-wage comparisons.

56. The chapter’s second section provides an account of the monetary framework debate. It first summarizes the current framework’s achievements and shortcomings, and then looks at traditional criteria for determining whether a peg or float fits Ukraine’s economic characteristics. After discussing the rather stringent policy requirements needed to make the peg work in Ukraine going forward, the section discusses alternative monetary policy strategies under a flexible exchange-rate regime: targeting a monetary aggregate and targeting inflation. The section make a case for inflation targeting as the framework that would be best suited for Ukraine to deal with future challenges. The section concludes by describing the state of the NBU’s preparations for inflation targeting and the further steps needed for the transition.

A. Ukraine’s Equilibrium Exchange Rate Reconsidered25

Recent External Sector Developments

57. Ukraine’s real effective exchange rate (REER) has been appreciating of late, after a long period without a trend. The recent appreciation is more pronounced when measured by the GDP deflator or unit labor costs, and has essentially returned the REER to levels not seen since prior to the 1998 financial crisis (Figure II.1) Looking at the source of shifts, with the exchange rate de facto pegged to the U.S. dollar since 2000, and Russia also limiting movements of its currency against the dollar, movements in the nominal effective exchange rate have been relatively small. The key factor of late has been high inflation in Ukraine relative to its trading partners.

Figure II.1.Ukraine: Real Effective Exchange Rate Developments

Sources: National Bank of Ukraine; and staff estimates.

58. The current account has recently swung widely. It rose to a surplus of over 10 percent of GDP in 2004, only to dive into deficit of 2 percent of GDP during the first half of 2006. Disaggregations are revealing:

  • The income and transfer balance components of the current account have improved steadily by about 1½ percent of GDP since 2000, largely due to higher remittances from Ukrainians working abroad.
  • Looking at the trade balance for goods and services (Figure II.2, left panel), shifts in trade prices have been important: the price of Ukraine’s major export, steel, rose markedly over the period. However, the dominant factor has been changes in volumes. In the period 2000–02, these can be well explained by real effective exchange-rate movements and by differences between partner import growth and Ukraine’s domestic demand growth. The fit is much poorer thereafter, raising questions about the quality of the trade data, an issue discussed in Appendix I.

Figure II.2Ukraine: Evolution of the Trade Balance and Current Account

  • Alternatively, tracing changes in the current account to shifts in savings and investment offers some clues about the driving forces behind more recent sharp shifts. The current account swings in 2004–05 were largely driven by shifts in domestic savings (Figure II.2, right panel). These in turn are likely linked to major changes in fiscal and incomes policies, which resulted in the transfer of large amounts of income from high savers (businesses) to heavy spenders (pensioners and wage earners).

59. The capital account has improved markedly since 2003 (Figure II.3). Direct investment has especially risen, reflecting large privatizations in 2004–05. Portfolio investment and bonds and medium-term loans have also ratcheted up since 2002, in the latter case reflecting both private and public sector borrowing. The inflows reflect benign world financial markets, but also appear to have an internal demand driven component: Ukraine’s banks are seeking funding abroad as they step up foreign-exchange lending to the private sector, particularly for mortgages and car loans (see Chapter III for discussion). Inflows have, however, been partly offset by short-term capital outflows. These are a residual, and thus include all unclassified items. They are thought to reflect the public’s demand for foreign-currency cash, as well as capital flight.

Figure II.3.Ukraine: Capital Flows

Source: National Bank of Ukraine.

60. Ukraine’s net international position should be improving, but the data tell another story. Recent current and capital account surpluses have been reflected in a build-up of foreign-exchange reserves, to almost $19 billion (4 ½ months of import cover) by end-2005. Moreover, the IMF’s World Economic Outlook classifies Ukraine as one of the few transition economies that have a net external credit position (i.e., positive cumulative current account balances since 1992). However, Ukraine’s officially measured international investment position (IIP) remains sharply negative, and has been rising of late (Figure II.4). The problem appears to stem from capital flight and foreign-currency cash accumulation, which lead to build up of external assets that are not well captured in IIP measurements. Appendix I reviews problems with Ukraine’s IIP data.

Figure II.4.Ukraine: Net International Investment Position

Source: National Bank of Ukraine.

The Macroeconomic Balance Approach

61. Implementing the macroeconomic balance approach requires three pieces of information (Figure II.5).26 First, an estimate of the level of the current account consistent with external balance over the medium term—the so-called current account (CA) norm. Second, an estimate of the level of the current account consistent with internal balance over the medium term—the underlying current account. And third, an estimate of the responsiveness of the current account to the real effective exchange rate over the medium term. To illustrate, for a country with current account deficit CAu (at real exchange rate Ru), and with a CA norm as depicted in Figure II.5, the real exchange rate would need to appreciate to R* to achieve both external and internal balance. Obviously, there are manifold uncertainties involved in making such estimates. Moreover, in a transition economy like Ukraine, the information needed to establish the equilibrium exchange rate may undergo significant changes over time, for example in the case of large terms-of-trade shocks or significant shifts in fiscal policy. Finally, it may or may not be the case that the estimated CA norm is sustainable, and thus estimates need to be cross-checked for their external sustainability implications.

Figure II.5.Medium-Run Real Exchange Rate Fundamentals

Estimating the current-account norm for Ukraine

62. The explanatory variables that go into the estimation of Ukraine’s current account norm are designed to capture the fundamental determinants of savings and investment (see Appendix II, Table II.A.2 for a listing of data sources and transformations):

  • Fiscal balance: In the absence of full Ricardian equivalence, an improvement in the fiscal balance should raise national saving and improve the current account (see Chinn, 2005 for empirical evidence on this point).
  • Foreign direct investment: Higher FDI inflows should allow a lower current account balance, either on sustainability grounds or by directly boosting imports.
  • Relative income levels: Economies lagging in per capita income have higher investment needs that should be reflected in a lower current account balance, while convergence-related growth can facilitate external debt servicing.
  • Demographic variables: A higher share of the economically inactive population would be expected to lower medium-term saving and the current account balance (see Higgins, 1998).
  • Energy balance: A sustained increase in energy prices represents an exogenous terms-of-trade shock. This could lower the medium-term current account balance for a net energy importer (e.g. Ukraine).27
  • Foreign-exchange reserve coverage: Under a pegged exchange-rate regime, higher reserves should allow a lower current account balance on sustainability grounds, but the direction of impact is less clear under a float. On the other hand, higher reserves could imply larger interest receipts, pointing to a higher current account balance.28
  • Other variables. Better governance and prospective EU membership could open the door to a lower risk premium and higher investment.

63. The estimated current account model covers the period 1994–2005 and uses annual data on 46 countries. The sample is about evenly split between industrial and transition economies; such broad coverage allows one to exploit substantial cross-country variation. The estimation was done in levels using standard least-squares panel techniques. The panel is estimated incorporating fixed effects: while fixed effects could end up capturing mostly cross-country variation in the data, they help mitigate possible omitted variable bias, which may be a problem in a pooled estimation.

64. The estimation results seem reasonable, and broadly in line with previous empirical work (TableII.1). The equation fit is good, with an adjusted R2 of 54 percent, while most parameters of interest are correctly signed, and statistically as well as economically significant. Improvements in the fiscal and energy balances of 1 percent of GDP produce 0.4 and 0.2 percent of GDP improvements in the CA norm. Increased FDI flows of 1 percent of GDP reduce it by ¼ percent of GDP. An extra month of reserve coverage raises the CA norm by 0.1 percent of GDP. Demographic variables turn out borderline significant, yet correctly signed and with non-negligible explanatory power: an increase in the old age dependency ratio and in population growth of 1 percent produce 0.15 and 0.1 percent of GDP deteriorations in the CA norm. However, the relative GDP variable is not statistically significant and is wrongly signed. This has been a typical finding in previous work, with the impact of relative GDP likely absorbed by the country-specific fixed effect due to its limited time variation over short samples.29

Table II.1.Model of Current Account Norm
VariableFixed effects estimation 1/
Fiscal balance0.370***
FDI-0.234***
Relative GDP-0.093
Dependency-0.142**
Pop. Growth-0.095*
Energy balance0.206**
Reserve coverage0.093*
Constant4.959
Adjusted R-squared0.539
Source: IMF staff estimates.

A *, **, ***, indicates significance at the 10, 5, and 1 percent levels, based on standard errors corrected for serial correlation.

Source: IMF staff estimates.

A *, **, ***, indicates significance at the 10, 5, and 1 percent levels, based on standard errors corrected for serial correlation.

65. Ukraine’s estimated current account norm has been consistently in deficit over the estimation period, and is projected to remain so in 2006 (Figure II.6, first panel). From 1996 to 2004, the deficit norm is estimated to have been moderate, without a trend, and ranging between 1¼ and 2 ½ percent of GDP (95 percent confidence intervals do not rule out that the norm could have been close to a balance, or as much as a 4 percent of GDP deficit during this period). During 2004–06, point estimates suggest that the deficit norm increased, peaking at over 4¼ percent of GDP. Looking at the evolution of the main explanatory variables between the 2002 norm trough and its 2005 peak, the main contributors to the increase are FDI, the fiscal position, the old-age dependency ratio, and the energy balance. Population growth and reserve coverage provided partial offsets.

Figure II.6.Ukraine: The Current-Account Norm

Source: Staff estimates.

66. A comparison of Ukraine’s norm and actual current accounts provides a rough indication of the degree of the hryvnia’s misalignment during the estimation period (Figure II.6, second panel). On this basis, it would appear that prior to the 1998 financial crisis the hryvnia was close to equilibrium and may even have become somewhat overvalued. In the crisis aftermath, a large gap opened between the actual and norm current accounts, indicating a substantial overshoot relative to equilibrium. In the context of the de facto peg at the new depreciated exchange rate, the currency appears to have remained substantially undervalued. Only in 2005–06 did a major correction of this misalignment take place.

Real exchange rate assessment

67. Quantitative estimates of real exchange-rate misalignment can be derived using the current account norm and by taking two further steps. First, an estimate of the underlying current account is needed. This is the actual current account adjusted for the impact of past real exchange-rate changes and corrected for the economy’s cyclical position (i.e. the deviation of actual from potential output in both Ukraine and its trading partners). Second, the change in the real exchange rate needed to bring the underlying current account into line with the current account norm must be estimated. This will depend on the responsiveness of import and export volumes to real exchange-rate changes.

68. Preliminary results from applying the macro balance approach suggest that the hryvnia remains somewhat undervalued. Staff projects a current account deficit of about 0.2 percent of GDP and an current account norm of 3 percent in 2006 (reflecting the return of FDI to more normal levels, reserve accumulation, and fiscal consolidation, which more than offset continued deterioration of the energy balance). We take the underlying current account to be the actual current account, since there are no well-defined cyclical fluctuations in Ukrainian time-series data, and since there is high uncertainty about lagged exchange-rate effects. Staff estimates for Ukraine suggest that a 1 percent real exchange-rate change will produce a 0.4 to 0.6 percent increase in the volume of exports and a -0.2 to -0.3 percent decrease for the volume of imports.30 At the same time, there is reason to assume that the terms of trade would not be affected by exchange-rate movements: the commodity composition of Ukraine’s trade suggest local currency pricing for exports and producer currency pricing for imports are reasonable assumptions. Thus, a real appreciation of between 3 and 15 percent would eliminate the projected 2.8 percent of GDP gap (Table II.2).

Table II.2.Ukraine: Real Exchange Rate Assessment
Elasticity 12006
CA norm: lower bound CA norm upper bound
(Percent of GDP)
Current-account norm-1.89-5.63
Projected current account-1.04-1.04
Change in current account to reach norm-0.85-4.59
of which:
From change in export volumes0.50-0.55-2.96
From change in import volumes-0.25-0.30-1.63
From change in export prices0.000.000.00
From change in import prices0.000.000.00
(Percent)
Change in REER needed to reach norm 2/-2.75-14.80
Source: Staff estimates.

Mid-point of elasticity range.

A negative sign signals an appreciation.

Source: Staff estimates.

Mid-point of elasticity range.

A negative sign signals an appreciation.

69. Looking forward, the extent of real exchange-rate misalignment in Ukraine would be expected to change over time. The CA norm should be affected by rising energy-import prices, or by rising FDI (in both cases shifting it further into deficit). At the same time, the actual current account balance could be affected by declining terms of trade, reflecting, for example, falling steel export prices (steel products account for some 40 percent of Ukraine’s exports and prices are presently significantly above their medium-term trend).

Cross check for debt sustainability

70. The estimated CA norm for Ukraine appears sustainable, especially once the level of foreign assets is accounted for. Assuming medium-term growth averaging around 5 percent and FDI around present levels, Ukraine’s gross external debt ratio should steadily fall for any norm within the estimated confidence interval (Table II.3). Levels of debt, even under the lower bound for the deficit norm, should fall back to about 31 percent of GDP. Using a measure of net foreign assets would provide for an even more favorable calculation, particularly if Ukraine’s implicit foreign asset accumulation since 2003 is taken into account, rather than Ukraine’s measured IIP.

Table II.3.Ukraine: Sustainability of Current-Account Norm
ScenarioCA norm

(deficit) 1/
Debt ratio

(2006)
Debt ratio

(2016)
(Percent of GDP)
Gross debt1.947.96.0
5.647.931.0
Net debt A 1/1.929.8-1.0
5.629.823.9
Net debt B 2/1.912.0-8.0
5.612.017.0
Source: Staff estimates.

Upper and lower bound of confidence interval.

Initial net debt equal to end-2005 international investment position, plus net current-account change in 2006.

3/ Initial net debt equal to end-2003 international investment position, plus net current-account change in 2004–6.
Source: Staff estimates.

Upper and lower bound of confidence interval.

Initial net debt equal to end-2005 international investment position, plus net current-account change in 2006.

3/ Initial net debt equal to end-2003 international investment position, plus net current-account change in 2004–6.

Assessing Longer-Term Real Exchange-Rate Trends

71. The macroeconomic balance approach of the previous section cannot address the question of where the real exchange rate is likely to settle in the longer run. This section briefly addresses this issue, based on two different approaches: a PPP-based methodology, which is fairly straightforward and widely used; and a dollar-wage-based approach.

PPP-based approach

72. PPP-based measures of the real exchange rate entail a comparison of actual GDP and GDP evaluated at purchasing power parity. The degree of misalignment is captured by the ratio between actual and PPP-adjusted GDP. The PPP-based approach has the advantage that it allows cross-country comparison of the degree of misalignment, since PPP-adjusted GDP is in theory comparable across countries.31

73. PPP-based measures suggest that the hryvnia is substantially more undervalued than the currencies of other transition economies (Figure II.7). In 2005, Ukraine’s PPP-based real exchange rate suggested a real undervaluation of the hryvnia of about 75 percent, considerably more than in neighboring countries such as Belarus, Russia and Georgia. This divergence appears to have held since 1995. As a point of comparison, the well-known Big Mac index, produced by the Economist magazine (May 22, 2006 issue) suggests an undervaluation of about 45 percent, in line with estimates for Russia and Moldova. Against this backdrop, Box II.1 discusses reasons why a PPP-based misalignment measure may be misleading for Ukraine.

Figure II.7.Transition Economies: Ratio of Actual to PPP-Adjusted GDP

Sources: IMF International Financial Statistics; and staff estimates.

74. Controlling for per-capita income can give a better picture. A country’s real exchange rate should appreciate as per capita GDP rises, since faster wage and productivity growth in the tradables sector should force non-tradable wages and prices to rise, and because income-elastic (and non-tradable) services should be in higher demand. A regression of the real exchange rate on PPP GDP per capita illustrates (Figure II.8). At Ukraine’s existing relative per capita income level, its degree of undervaluation would need to decline by only 9 percent to reach the estimated transition-economy path. Thereafter, its degree of undervaluation (as measured by the gap between actual and PPP-adjusted GDP) would be expected to decline by about 0.9 percent for every 1 percent improvement in per capita income. In fact, from 2000 to 2005, only 3⅓ percent of the undervaluation was eliminated, and Ukraine moved farther away from the transition path. This perhaps surprising result may reflect the data problems discussed in Box I.1

Figure II.8.PPP-Based Real Exchange Rate

Sources: IMF International Financial Statistics; and staff estimates.

Box I.1.Problems with PPP-Based Measures of GDP in Ukraine

PPP-based measures of GDP are constructed by revaluing a country’s output at international reference prices. Between revaluation exercises, updated estimates are constructed assuming that the production structure remains the same (that is, by using real growth in the particular economy, along with growth in the reference prices). Problems can arise because: (i) the prices for domestic output may be initially mismeasured; and (ii) the basket of goods may initially be or become incorrect (with too little weight placed on goods with prices closer to world prices). The farther away from the revaluation exercise one gets, the more likely a problem can arise, as economic changes accumulate.

Looking at Ukraine since 2000, the date of the last revaluation exercise, a key problem seems to be the production basket. Specifically, since 2000, the share of unliberalized agriculture in GDP has declined by 3 percentage points, while the share of the liberalized trade and transport sectors have risen by almost 6 percent of GDP. Industry’s share has been stable, but the growth rates in industries subject to heavy price control—energy (coal) mining and communal services—have cumulatively amounted to only 15 and 8½ percent. Meanwhile the growth rates in the chemical, metals and machine building industries, large export-oriented sectors where world prices are the norm, have amounted to 88, 67 and 184 percent respectively.

The changes in Ukraine’s production basket imply that PPP-based measures of GDP could be overstated. With more of GDP than assumed at prices close to world levels, the PPP mark-up becomes too high.

Dollar-wage approach

75. Dollar-wage-based measures of the real exchange rate rely on a comparison of actual with estimated equilibrium real wages. The equilibrium level of dollar wages is specified as a function of a number of productivity variables: the country’s per capita income, the capital stock, and its level of development. If the actual dollar wage falls short of the equilibrium level, this suggests that the wage rate was “overly” competitive, implying by extension an exchange rate undervalued in real terms, with the degree of undervaluation captured by the ratio of the actual to the equilibrium dollar wage. The dollar-wage approach was the basis for the last comprehensive IMF assessment of Ukraine’s competitiveness.32

76. The simple relationship between dollar wages and per capita income illustrates the broader approach, and provides a point of comparison with PPP-based estimates (Figure II.9). In this measurement, Ukraine’s dollar wages are only 6¼ percent of the EU25 average, suggesting massive real exchange-rate undervaluation from a longer term perspective. However, in both 2000 and 2004, Ukraine’s dollar wages have been broadly in line with the estimated transition economy convergence path (which entails dollar wage growth, and real appreciation, of about 1¾ percent per 1 percent increase in PPP-adjusted per capita GDP). By this perspective, the valuation of the hryvnia has been and remains consistent with past transition economy experience and, in this narrow sense, appropriate.

Figure II.9.Dollar-Wage Real Exchange Rate

Sources: ILO; and staff estimates.

B. What Monetary Policy Framework Fits Ukraine?33

Ukraine’s Present Monetary Policy Framework

77. The exchange rate to the U.S. dollar has served as the nominal anchor in Ukraine. Since June 2000, the official exchange rate has moved only very gradually from Hrv/US$ 5.45 to Hrv/US$ 5.05 with the biggest change occurring in April 2005, when the hryvnia appreciated by 4.6 percent. For most of the period, the rate was left unchanged (Figure II.10). The role of the exchange rate as a monetary anchor goes back even further to mid-1996, when the NBU first introduced a band around the U.S. dollar but had to widen and shift it upwards during the 1998–2000 financial crisis. Thus, except for a brief period of a (forced) float in late 1999, monetary policy has targeted the exchange rate for more than a decade.

Figure II.10.Ukraine: The Hrynia/U.S. Dollar Exchange Rate, 2000–06

Sources: National Bank of Ukraine; Bloomberg; and staff estimates.

1/ Determined daily by the NBU, but not necessarily identical with the rate at which the NBU intervenes in the market.

2// Average rate for interbank transactions.

78. Despite slightly greater nominal exchange-rate flexibility since April 2005, the exchange-rate regime still exhibits the characteristics of a de facto peg. Since the step revaluation in April 2005, the NBU has held the official rate constant at Hrv/US$ 5.05, except for a short period, from May 31 to July 18, 2005, in which it set the rate at Hrv/US$ 5.055. The interbank exchange rate was allowed to deviate from the official rate by up to 2 percent until September 2005, when the NBU started to target a 1 percent corridor of Hrv/US$ 5.00–5.06 (Figure II.10). Larger fluctuations occurred only sporadically. With an official and de facto corridor of less than 2 percent, the current exchange-rate regime is considered a de facto peg under the IMF’s official classification of exchange-rate regimes.34

79. Officially, the peg has been complemented by targets for monetary aggregates, but, in practice, the peg has taken precedence over those targets. Projected paths for base and broad money have been laid out in the NBU’s annual Monetary Policy Guidelines (Table II.4). Since for most of the time, base money growth was driven by the NBU’s foreign exchange interventions and the government’s financing requirements—with little active liquidity management by the NBU (Figure II.11, Table II.5)35—base money targets were frequently missed. But despite, at times, deviations from target, the NBU was able to contain inflation to below 15 percent as remonetization was much faster and stronger than predicted.

Table II.4.Ukraine: Base Money Targets(Percent change)
Original

Targets 1/
Revised

Targets 2/
Outcome
2005 3/20–2638–4353.9
2004 4/26–3251–5634.1
200317–2038–4230.1
200211–1332–3633.6
20011218–1937.4
Sources: National Bank of Ukraine; and IMF International Financial Statistics.

Set in Monetary Policy Guidelines in Sept. of previous year.

Revised in Sept. of current year.

Revision in Nov. 2005 to 50–55 percent.

Revision in Dec. 2004 to 34–40 percent.

Sources: National Bank of Ukraine; and IMF International Financial Statistics.

Set in Monetary Policy Guidelines in Sept. of previous year.

Revised in Sept. of current year.

Revision in Nov. 2005 to 50–55 percent.

Revision in Dec. 2004 to 34–40 percent.

Figure II.11.Ukraine: Contributions to Base Money Growth

Sources: National Bank of Ukraine; and staff estimates.

Table II.5.Ukraine: Variance Composition of Base Money, 2000–06 (Percent)
Contribution
Net international reserves139.7
Net claims on government-39.7
Net claims on banks-2.9
Other items net2.9
Total100.0

80. Ukraine’s inflation has been volatile and well above that of more advanced transition economies (Figure II.12). CPI inflation stayed below 15 percent during the past 5½ years, a success when compared to the hyperinflation of the early nineties and the reemergence of inflation to some 30 percent after the financial crisis. But when compared to other transition economies, the record is less favorable. At an average of 9.3 percent over the past 3½ years, inflation in Ukraine was nearly double that of Central and Eastern European countries, though similar to many other CIS countries. Inflation in Ukraine was also highly volatile, fluctuating between zero and 15 percent—a range exceeded by only a few other transition countries.

Figure II.12.Transition Economies: Inflation Indicators, 1998–2005

Sources: National Bank of Ukraine; State Statistics Committee; and staff estimates.

1/ CIS countries excluding Tajikistan, Turkmenistan, and Uzbekistan.

The De Facto Peg: Why it May Have Fit Ukraine in the Past

81. Ukraine has developed a strong affinity for its pegged exchange-rate regime for several reasons. Heartened by a successful nominal stabilization after the 1998 crisis, the public has viewed a stable exchange rate as the guarantor of internal and external stability—inflation of below 15 percent compared favorably to the early transition years, and levels of foreign reserves surged to 4½ months of imports. Against this backdrop, many fear that allowing greater fluctuations of the exchange rate would risk undermining the hard-earned trust in the currency with negative balance sheet effects on the economy. In addition to this largely psychological argument, many have questioned Ukraine’s readiness to operate under more exchange-rate flexibility: financial markets are underdeveloped, modeling and forecasting capacities for inflation are still being established, and experience with more active monetary policy has been limited.

82. Ukraine’s choice of the exchange-rate regime also mirrors that of many other CIS countries. Most CIS countries have, over the last few years, limited the fluctuations of their currencies against the U.S. dollar to far less than 5 percent (Table II.6). Experiments with greater exchange-rate fluctuations have been brief and, even though many regimes are still classified as “managed floats” by the IMF, they are de facto “tightly managed floats.” Under an IMF classification scheme currently under consideration, they would be classified accordingly. This has left the region without a point of reference in moving to a more flexible exchange and monetary regime. The notable exception is Armenia, which over the past 12 months, has allowed for the greatest exchange-rate variability in the region and on July 1, 2006 announced the adoption of inflation targeting.

Table II.6.CIS Countries: Indicators for the Exchange-Rate Regime
Exchange rate regime 1/Reference currency 2/Maximum range (percent) 3/Maximum daily

volatlity (percent) 4/
Number of daily changes > 1 %Variation Coefficient of Official Exchange Rate (Average 2001–06) 5/
ArmeniaIndependent floatRussian ruble9.61.890.11
AzerbaijanPegU.S. dollar6.00.900.03
BelarusPegU.S. dollar0.60.200.19
GeorgiaManaged floatU.S. dollar4.01.220.08
KazakhstanManaged floatRussian ruble10.71.120.07
Kyrgyz RepublicManaged floatU.S. dollar4.11.110.07
MoldovaManaged floatU.S. dollar8.10.800.05
RussiaManaged floatU.S. dollar8.61.010.05
TajikistanManaged floatU.S. dollar6.00.500.11
UkrainePegU.S. dollar4.01.010.02
Memorandum item
Euro areaIndependent floatU.S. dollar11.02.0210.15
Sources: IMF International Financial Statistics; IMF Annual Exchange Rate Arrangements and Restrictions; and staff calculations.

Classification in IMF Annual Exchange Rate Arrangements and Restrictions (2006).

Currency to which the variance is smallest.

Percent deviation of maximum interbank rate from minimum interbank rate from August 2005-July 2006 (using daily interbank rates).

Maximum day-to-day change in interbank exchange rate from August 2005-July 2006 (in percent).

Classification in Standard deviation divided by mean.

Sources: IMF International Financial Statistics; IMF Annual Exchange Rate Arrangements and Restrictions; and staff calculations.

Classification in IMF Annual Exchange Rate Arrangements and Restrictions (2006).

Currency to which the variance is smallest.

Percent deviation of maximum interbank rate from minimum interbank rate from August 2005-July 2006 (using daily interbank rates).

Maximum day-to-day change in interbank exchange rate from August 2005-July 2006 (in percent).

Classification in Standard deviation divided by mean.

83. This preference is partly backed up by traditional arguments on the choice of optimal exchange-rate regimes. Husain (2006) proposes a template for assessing whether or not a country’s economic and financial characteristics make it an appropriate candidate for a pegged or a floating regime (Box II.2). The template uses various indicators that have been identified in the literature as key potential determinants of exchange-rate regime choice: economic diversification, trade integration, financial integration, macroeconomic stability, credibility, and “fear-of-floating” type effects.

84. When applied to Ukraine’s past economic and financial characteristics, the indicators slightly favor a peg over a float for Ukraine (Box II.2):

  • Economic diversification: The volatility of the terms of trade has been significant and the correlation between Ukraine’s business cycle and world prices for commodities has been strong. This would argue for a more flexible regime. The share of commodities in exports, however, puts Ukraine close to the average of countries and therefore did not strongly favor either regime.

Box II.2.A “Scoreboard” for Ukraine’s Exchange-Rate Regime

Using a sample of 51 countries, Husain assigns scores for the suitability of a peg by applying quantitative measures to the literature’s standard arguments. For each of the six main arguments, he uses two to three indicators. If a country ranks in the top 10 percentile, this is taken as a strong case for a peg, a ranking in the top 20 percentile as a somewhat less strong case for a peg. Rankings in the bottom 10 and 20 percentiles are seen as strong or somewhat less strong cases for a float.

Applying this methodology and updating some of Husain’s indicators provides the following scoring for Ukraine. The findings and comparisons with other transition economies are only indicative though, as they have not been updated and the rating depends on Ukraine’s relative position to other countries.

Scorecard for Exchange-Rate Regimes in CIS Countries
UkraineUkraine

Updated 1/
GeorgiaKazakhstanKyrgyz

Republic
Russia
Economic diversification342433
Trade integration133233
Financial integration231113
Macroeconomic stabilization222334
Credibility223221
Fear of floating effects223322
Total2.02.72.32.52.32.7
1= strong case for peg; 2= case for peg; 3=neutral; 4=case against peg; 5=strong case against peg
Sources: Husain (2006); and staff estimates.

Data has been updated to mid-2006.

Sources: Husain (2006); and staff estimates.

Data has been updated to mid-2006.

  • Trade integration: The data on openness to trade and trade pattern concentration suggest that there is no strong case for either a peg or a float. But Ukraine’s relatively weak cyclical synchronization over the past six years with its major trading partner Russia, argued against a peg.
  • Financial integration: Overall, no strong preference for either exchange-rate regime can be derived based on Ukraine’s past financial integration. Its access to international financing argued for a float, while the still relatively small (though rapidly rising) stock market argued for a peg. The level of financial intermediation in the past put Ukraine right in the middle of both regimes.
  • Macroeconomic stability: A relative low degree of capital mobility, highly volatile money velocity, and the ratio of terms-of-trade to velocity all favored pegging.
  • Credibility: Ukraine’s relatively low monetary credibility, as measured by the period the country was below an inflation threshold of 8–10 percent, also supported the case for pegging.
  • Fear-of-floating effects: While the level of dollarization did not clearly argue for or against pegging, the other two attributes assumed to be linked to fear of floating (correlation between real activity and exchange rate; and pass-through of exchange-rate changes to inflation) argued for pegging.

85. While these traditional arguments make a case why a peg may have been an acceptable fit for Ukraine in the past, they also suffer from circularity. In many ways, the existence of the peg created some of the conditions that, in the above analysis, are used to argue for the use of the peg. For example, under the peg and in light of positive terms-of-trade shocks, it was difficult to keep inflation consistently in mid-single digits. This lack of internal stability or credibility of monetary policy is used as an argument for a peg. Two other examples are the underdevelopment of financial markets and the fear of float, which have both, to a large degree, been caused by the peg itself.

Keeping the Peg: What it Would Take to Make it Fit for Future Challenges

86. Going forward, new challenges could make operating a peg even more complicated. In the past, the monetary framework has revealed weaknesses in terms of internal and financial stability. Going forward, it will be even more challenging to meet these objectives as the economy becomes more integrated into the world economy and thus more susceptible to capital flow and terms-of-trade shocks. A sharp terms-of-trade deterioration could also pose a serious challenge to external stability, which has not been a concern in the past. Without the nominal exchange rate as an adjustment tool, goods and labor markets would need to adjust flexibly in response to shocks to maintain internal and external stability.

87. Transition countries that have successfully operated a peg and maintained internal and external stability usually have relatively small, open, and flexible economies (Table II.7). The Baltics and Bulgaria are examples where a very rigid exchange rate regime was supplemented by rather flexible labor and goods markets, as well as disciplined fiscal frameworks.36 This helped to contain inflation at 3½ percent on average over the past five years and to maintain competitiveness. In contrast, countries with originally less flexible institutions and policies chose to let the exchange rate play a greater role as an adjustment mechanism, namely the Czech Republic, Hungary, and Poland, which held inflation in the low mid-single digits, and more recently, Armenia and Romania. A number of countries that have allowed only a small degree of exchange-rate flexibility are facing the regime’s challenges. For example, Croatia and the Slovak Republic’s relatively rigid labor markets have contributed to high unemployment. In other countries, such as Moldova, Russia, and Serbia and Montenegro, inflation has remained relatively high, mirroring the inflation control problem under a tightly-managed float.

Table II.7.Transition Economies: De Facto Exchange-Rate Arrangements and Anchors of Monetary Policy

(as of July 31, 2006) 1/

Monetary Policy Framework
Exchange rate regimeExchange rate

anchor
Monetary

aggregate target
IMF supported or

other monetary

program
Inflation targeting

framework
Other
Currency boardBosnia & Herzegovina (1.2 %)

Bulgaria (4.5 %)

Estonia 2/ (2.8 %)

Lithuania 2/ (0.9 %)
PegBelarus 4/ (18.9 %)

Latvia (5.3 %)

Macedonia, FYR (0.7 %)

Turkmenistan (7.4 %)

Ukraine (9.3 %)
Peg within horizontal bandsSlovenia 3/ (3.9 %)Hungary 3/ (5.0 %)
Crawling peg
Exchange rates within crawling bands
Tightly managed floatCroatia (2.4 %)

Georgia (6.3 %)

Moldova (12.1 %)

Serbia and Montenegro (12.4 %)

Tajikistan (10.2 %)
Kazakhstan (7.0 %)

Russia (12.4 %)
Managed floatAzerbaijan (6.2 %)

Kyrgyz Republic (3.8
Czech Republic (1.6 %)

Romania (12.1 %)
Slovak Republic (6.3 %)

Uzbekistan (14.9 %)
Independent floatAlbania (2.6 %)Armenia (4.1 %)

Poland (2.1 %)
Source: IMF Annual Report on Exchange Arrangements and Exchange Restrictions.

Average annual CPI inflation from 2003–05 in brackets.

The country participates in the ERM II mechanism of the European monetary system.

The bandwidth is +/- 15 percent.

The band width is adjusted frequently.

Source: IMF Annual Report on Exchange Arrangements and Exchange Restrictions.

Average annual CPI inflation from 2003–05 in brackets.

The country participates in the ERM II mechanism of the European monetary system.

The bandwidth is +/- 15 percent.

The band width is adjusted frequently.

88. Key requirements “for making a peg work” in Ukraine would be a fiscal policy oriented toward price stability and a high degree of wage and price flexibility. One illustration of the degree of flexibility needed is the adverse medium-term terms-of-trade shock under the staffs baseline scenario (Table II.8). The cumulative terms-of-trade shock is some 20 percent. To maintain external sustainability, staff estimates that the real effective exchange rate needs to remain broadly unchanged over the medium term, in contrast to an estimated 30 percent real appreciation if the terms of trade were to stay constant.37 Under a flexible exchange rate, adjustment is shared between the exchange rate, monetary, and incomes policies. Under a peg, relative price adjustment would have to come entirely through domestic wages and prices, to which tighter fiscal and incomes policies would need to contribute. This would help reduce aggregate demand and keep the current account sustainable.

89. Table III.8 summarizes the estimated differences in macroeconomic outcomes under fixed and flexible exchange-rate regimes. It assumes—as an indicator of external sustainability—that the NBU aims to maintain the same level of foreign exchange reserves under both exchange-rate regimes. Under the peg, very tight fiscal and income policies would be needed, which would cause real GDP growth and inflation to significantly undershoot medium-term norms. The estimated cumulated output loss, compared to a scenario in which the exchange rate is used as an adjustment mechanism, is 8 percent. Inflation would have to decelerate much faster under a peg than under a float (and would be cumulatively 20 percent lower for the five year period).

Table II.8.Ukraine: Macroeconomic Framework under an Terms-of-Trade Shock—Outcomes under a Flexible and Fixed Exchange Rate %1/
20072008200920102011Cumulative

Change 2007–11
Differences between a Flexible and Fixed Exchange Rate
Output and prices
Real GDP growth (Percent change)2.52.31.50.7-0.27.8
Consumer prices (Percent change; end of period)4.74.52.82.31.819.8
Consumer prices (Percent change; average)2.44.63.62.52.019.6
Wages
Minimum wage (Hryvnias per month; end of period)75.0103.0116.0148.0184.0
Nominal monthly wages (Percent change; average)8.97.54.32.73.836.2
Real monthly wages (Percent change; average)5.92.70.60.11.711.5
Public finance (Percent of GDP)
Cash balance-1.8-2.6-2.5-2.0-1.5
Public debt (end of period)1.83.54.96.26.6
External sector
Current account balance (Percent of GDP)0.30.0-0.5-0.6-0.8
Gross official reserves (Months of imports of goods and services)0.20.10.20.10.0
Real effective rate (Percent change) %2/-4.1-3.01.10.22.5-3.2
Goods and services terms of trade (Percent change)0.00.00.00.00.00.0
Sources: Ukrainian authorities; and staff estimates and projections.

Assumes a cumulative terms-of-trade deterioration of 18 percent between 2007–11.

Period averages; (+) represents real appreciation; based on GDP deflator and INS trade weights (1999–2001).

Sources: Ukrainian authorities; and staff estimates and projections.

Assumes a cumulative terms-of-trade deterioration of 18 percent between 2007–11.

Period averages; (+) represents real appreciation; based on GDP deflator and INS trade weights (1999–2001).

90. In Ukraine, achieving such a high flexibility in macroeconomic policies would be challenging. Since the government’s setting of minimum wages and pensions is closely linked to those in the private sector, its policies would largely determine the lower boundaries for wage and pension growth in the economy. Going forward, political pressures, in light of differing political priorities and upcoming elections, would make steering a tight course, as outlined under the shock scenario, very difficult.

91. Apart from the unlikely availability of such complementary policies, maintaining a peg would also carry forward the difficulties in achieving internal and financial stability.

  • Even in the absence of term-of-trade shocks, it may be difficult to achieve low and stable inflation. As shown in Chapter II.A, the hryvnia, like the currencies of other transition economies, is expected appreciate in real terms over the medium term as structural reforms take place. In the case of a pegged nominal exchange rate, the real appreciation would be achieved through relatively high inflation.
  • It could stifle financial market development and make it difficult to reverse dollarization, thus fostering the buildup of financial sector risks (see Chapter III). As long as the volatility of inflation remains relatively high compared to the volatility of the real exchange rate, there would not be much incentive to de-dollarize and improve bank risk management. Moreover, the de facto exchange-rate guarantee would continue to favor the issuance of securities in foreign currency over hryvnia, thereby stifling the development of a domestic securities market, which is needed to improve the interest-rate channel of the transmission mechanism.

What Monetary Framework Is a Better Fit for Ukraine? The Options

92. Shifting to greater exchange-rate flexibility raises the question how monetary policy should operate. When allowing the hryvnia to fluctuate more freely, two options for a new monetary strategy can be considered: monetary targeting and inflation targeting. This section compares the policy requirements of each and makes a case for inflation targeting as the framework that would be best suited for Ukraine to achieve internal, external, and financial stability.

Monetary targeting

93. A key requirement for monetary targeting is a strong short-term link between money and inflation—a condition presently not in place in Ukraine. Over the past years, money demand in Ukraine in the short run fluctuated significantly and, over the medium term followed an unexpected strong upward trend—underestimated by most forecasters, including the IMF. These two factors together contributed to large forecast errors in monetary aggregates (Table II.4). Empirical studies could not identify a strong and stable short-term relationship between money and inflation, even though they did find an impact of money on the inflation process.38 The main reasons for the lack of money-demand stability are the ongoing structural changes of the economy, including continuing catch-up effects, adjustment of relative prices, and introduction of new financial instruments, as well as the effect of dollarization (including cash holdings in U.S. dollars) and the impact from financial “round-tripping” operations of corporates and bank conglomerates that appear to have inflated monetary aggregates.39 Since these processes are expected to continue for the foreseeable future, it is unlikely that a money-demand function will emerge that could serve as the operational foundation for money targeting.

94. The missing close short-term link between money and inflation was also the cause for disappointing experiences with money targeting in other transition economies. Money demand has been unstable in most transition economies (Figure II.13 shows large shifts in velocity), and led to the abandonment of the regime by those countries which attempted it. In particular, the Czech Republic and Poland gave up monetary targeting and moved to inflation targeting in 1997/98. Many other transition economies have used money targets as complements to their exchange-rate anchor, but have given prominence to the latter when targets conflicted. Examples, in addition to Ukraine, are Kazakhstan, Kyrgyz Republic, Serbia and Montenegro, and the Slovak Republic. Currently, the only country that officially follows a monetary targeting framework is Albania. There an intermediate target is set for items of the central bank balance sheet (net domestic assets, net credit to the government, and net international reserves). But even in Albania’s case, operating the framework is challenging, given the economy’s structural changes and the fact that the intermediate targets do not set an upper bound for money growth (because only a floor for net international reserves is defined). As a consequence, the targets have been revised frequently and the Bank of Albania is now preparing to move to inflation targeting.

Figure II.13.Transition Economies: Trends in Velocity of Broad Money 1/

Sources: IMF World Economic Outlook; and staff calculations.

1/ Nominal GDP divided by end-year broad money.

95. Nevertheless, money aggregates can serve an important indicator function for monetary policy, in particular in transitioning to inflation targeting. Even though not well-suited as a target variable with a one year horizon, money could play a role as a leading indicator for the NBU’s monetary-policy decisions. Using it as an indicator rather than a target would allow a more flexible treatment of instabilities in the relationship between money and inflation, as it does not bind a central bank to a target path nor would it undermine its credibility in case of repeated target adjustments or misses. In particular, in the transition period to inflation targeting, it could fill a gap, until the interest-rate transmission mechanism has strengthened, by providing easily monitorable guidance. And even after the adoption of full-fledged inflation targeting, many central banks include monetary aggregates in their information set that underlies their inflation projections and policy decisions.

Inflation targeting

96. Shifting to an inflation-targeting regime seems to be the most promising option for Ukraine to deal with current and future monetary-policy challenges. First, monetary policy’s room for maneuver would increase under a more flexible exchange rate in a way that should help it to regain better control over inflation. An inflation target would provide a new transparent monetary-policy anchor. Second, the greater variability of the real exchange rate should also serve as disincentive for dollarization and excessive risk-taking of the banking sector. Third, greater exchange-rate flexibility would facilitate adjustments to terms-of-trade shocks and buffer the needed fiscal and wage responses. And fourth, changes in Ukraine’s economic structure, such as the development of financial markets and capital-account liberalization, would shift financial characteristics in favor of a float.

97. Inflation targeting has been successfully adopted by a number of transition and emerging market economies with similar policy challenges. An analysis of the starting conditions of 13 full-fledged inflation targeting emerging market countries (IMF, 2005) showed that they were far from ideal but improved over time (Table II.9). Moreover, the recent examples of Armenia, Colombia, Peru, Romania, and Turkey indicate that it is indeed feasible to shift to inflation targeting even under difficult circumstances, such as high levels of dollarization and underdeveloped financial markets. The starting conditions of these countries, which are summarized in Table II.10, were not better than those in Ukraine. The effects that the adoption of inflation targeting by emerging market countries had on their economies was studied by IMF (2005) and Roger and Stone (2005). They find that the level and the volatility of inflation dropped significantly once emerging market countries introduced the new regime. Non-inflation targeting emerging market economies were also able to lower inflation and its fluctuations, but they were clearly outperformed by the inflation targeters. These inflation stabilization successes did not come at the expense of real output stabilization, in which inflation targeters also outperformed their peers.

Table II.9.Initial Conditions for Emerging Market Inflation Targeters 1/(Index; 0 = poor; 1 = ideal)
Pre-adoption of

inflation targeting
CurrentPre-adoption of

inflation targeting
Current
Technical infrastructure0.290.97Institutional independence0.590.72
Data availability0.630.92Fiscal obligation0.771.00
Systematic forecast process0.101.00Operational independence0.810.96
Models capable of conditional forecasts0.131.00Central bank legal mandate0.500.62
Governor’s job security0.850.85
Financial system health0.410.48Fiscal balance in percent of GDP0.480.47
Bank regulatory capital to risk-weighted assets0.751.00Public debt in percent of GDP0.470.47
Stock market capitalization to GDP0.160.21Central bank independence0.260.64
Private bond market capitalization to GDP0.100.07
Stock market turnover ratio0.290.22Economic structure0.360.46
Currency mismatch0.920.96Exchange rate pass-through0.230.44
Maturity of bonds0.230.43Sensitivity to commodity prices0.350.42
Extent of dollarization0.690.75
Trade openness0.180.21
Source: IMF World Economic Outlook, September 2005, p.176.

Includes Brasil, Chile, Colombia, Czech Republic, Hungary, Israel, Korea, Mexico, Peru, Philippines, Poland, South Africa, and Thailand.

Source: IMF World Economic Outlook, September 2005, p.176.

Includes Brasil, Chile, Colombia, Czech Republic, Hungary, Israel, Korea, Mexico, Peru, Philippines, Poland, South Africa, and Thailand.

Table II.10.Selected Inflation Targeting Countries: Macroeconomic Indicators at the Time of the Regime Shift
Year before adoption of Inflation TargetingUkraineArmeniaColumbiaCzech

Republic
HungaryMexicoPeruPolandRomaniaTurkey 1/
2005200519981997200020002001199820042004
Inflation
Real GDP growth (percent change)2.613.90.6-0.76.06.60.25.08.48.9
CPI (period average; percent change)13.50.618.78.49.89.52.011.811.98.6
CPI (end of period; percent change)10.3-0.216.710.08.9-0.18.69.39.4
Monetary and financial variables
Broad money (percent of GDP)45.716.435.666.046.245.026.036.726.242.8
Credit to the private sector (percent of GDP)35.68.123.570.032.017.224.122.517.519.8
Foreign currency deposits as percent of total broad money23.638.516.211.419.72.967.04.041.2
Foreign currency deposits as percent of total deposits 2/34.263.625.612.623.31.980.015.239.544.0
Financial system health
Capital adequacy ratio (percent)15.033.710.39.513.713.812.811.718.828.8
Stock market capitalization (percent of GDP)28.70.717.224.129.020.420.213.017.621.4
Bank assets (percent of GDP) 3/52.320.238.5106.860.053.143.158.037.158.3
Domestic interest rate spread (percentage points) 3/8.312.24.45.53.18.710.56.314.19.6
Fiscal variables
Fiscal balance (percent of GDP)-2.4-2.6-4.0-1.2-2.7-3.7-2.6-2.5-1.0-8.1
Total public debt (percent of GDP)19.425.722.110.555.623.244.342.923.163.5
Domestic debt (percent of GDP)4.82.110.67.830.312.39.65.346.0
External debt (percent of GDP)14.623.611.52.725.310.934.717.817.5
Indicators of economic structure
Nominal GDP (US $ billions)82.94.998.457.447.0580.553.6172.071.4302.6
Nominal GDP per capita (PPP)1,7471,1402,5055,5454,6005,9352,0474,4413,4644,289
Current account balance (percent of GDP)3.1-3.3-4.9-6.2-8.5-3.2-2.2-4.0-8.5-5.2
Trade openness (percent of GDP) 4/85.980.231.3108.7151.642.533.555.486.666.8
Sources: IMF International Financial Statistics; IMF World Economic Outlook; Bloomberg; and staff calculations.

Public external and domestic debt in percent of GNP.

For Poland foreign currency liabilities in percent of total bank liabilities.

Data for 2002 for Czech Republic.

The sum of exports and imports of goods and services in percent of GDP.

Sources: IMF International Financial Statistics; IMF World Economic Outlook; Bloomberg; and staff calculations.

Public external and domestic debt in percent of GNP.

For Poland foreign currency liabilities in percent of total bank liabilities.

Data for 2002 for Czech Republic.

The sum of exports and imports of goods and services in percent of GDP.

Adopting Inflation Targeting: Making the Transition Work

98. The NBU has committed to gradually move to inflation targeting. In its Monetary Policy Guidelines from September 2005, the NBU articulated its plan to eventually move to inflation targeting and in January 2006, it adopted an Action Plan that outlines a number of operational steps for the preparatory work, many of which have already been taken (Table II.11). However, an official timeframe for taking the key policy decision on greater exchange-rate flexibility does not yet exist.

Table II.11.Ukraine: Moving to Inflation Targeting: Progress Report
PreconditionsKey AchievementsFurther Steps Needed 1/
Mandate to pursue price stability
  • Drafted a Memorandum of Understanding (MoU) that sets out the roles of the NBU and the government in inflation targeting.
  • Submitted draft amendments to the NBU Law that would increase the independence of the NBU.
  • Agree with the government on the MoU.
  • Adopt further amendments to the NBU Act that would provide the NBU with a clearer mandate to achieve price stability and allow for greater NBU independence.
Exchange-rate flexibility
  • Allowed the hryvnia/U.S. dollar rate to fluctuate in a 1 percent band.
  • Draw up concrete operational transition plan, including on the changing role of the exchange rate.
  • Gradually allow greater exchange-rate flexibility.
Monetary-policy instruments
  • Identify key policy rate; assess possibility to introduce an interest rate corridor.
  • Actively steer a short-term interest rate.
  • Apply monetary instruments in a consistent, transparent, and market-orient manner.
Capacity to model and forecast
  • Developed a small macroeconomic model.
  • Compiled an inflation report, including inflation forecasts.
  • Published a business survey, including on inflation expectations.
  • Improve macroeconomic modeling.
  • Develop core inflation indicators.
Communication of monetary policy
  • Issued a brochure and an analytical paper on inflation targeting and the transmission mechanism of monetary policy.
  • Issued a monetary-policy report.
  • Held seminars on inflation targeting within the NBU and for commercial banks.
  • Modernize the NBU website.
  • Conduct press conferences and issue press releases on inflation targeting.
  • Issue an Inflation Report.
Financial-market development
  • Liberalized the foreign-exchange market in August 2005.
  • Government to abolish the foreign-exchange turnover tax.
  • Develop benchmark government securities.
  • Simplify procedures for working in the foreign-exchange market.
Banking-sector stability
  • Developed a strategy for the medium-term development of the banking sector and supervision.
  • Gradually implement more risk-based supervision.
  • Further tighten banking supervision, regulation, and legislation.

Most of the steps listed here are part of the NBU’s Action Plan toward adopting inflation targeting, which was adopted by the NBU Board in January 2006.

Most of the steps listed here are part of the NBU’s Action Plan toward adopting inflation targeting, which was adopted by the NBU Board in January 2006.

99. Nevertheless, a number of challenges and outstanding operational issues still need to be addressed and this calls for a gradual transition. The authorities’ main concern relates to the risks for the economy from greater exchange-rate flexibility as well as the feasibility to operate inflation targeting given underdeveloped financial markets and the weak transmission mechanism. How these and other issues can possibly be addressed is briefly summarized below.

  • The role of the exchange rate during transition: Staffs’ analysis (see Chapter III.B) suggests that the banking sector is in a position to deal with more exchange-rate flexibility, though very sharp and large adjustments could be risky. The likelihood for such sharp adjustments could be minimized by various policy decisions: First, the NBU should allow the exchange rate to start fluctuating in a gradually widening band, as practiced by many other emerging market economies that have moved to inflation targeting. Since the exchange rate pass-through to prices is relatively high, it can be expected that the exchange rate will continue to play a role even once Ukraine has moved to full-fledged inflation targeting (Ho and McCauley, 2003). And second, starting the transition swiftly would allow the NBU to operate from a position of strength. Its relatively high level of foreign reserves could serve as a buffer for potential excessive fluctuations. In the meantime, the excessive foreign exchange turnover tax of currently 1.3 percent should be eliminated to develop the foreign exchange market.

Box II.3.IMF Technical Assistance Provided to Ukraine in Preparation for Inflation Targeting

The IMF has actively supported the NBU’s transition preparations. The efforts over the past three years included technical assistance and sharing of other country experiences. Topics ranged from inflation forecasting, monetary and foreign-currency operations, development of government securities markets, and central bank communication strategy.

IMF Technical Assistance to Ukraine on Inflation Targeting 1/
  • Overview on Preconditions for an Inflation Targeting Framework (TA mission April 2004; seminar held at the NBU in November 2004).
  • Modeling and Operating Inflation Targeting (Resident Advisor to the NBU, Nov. 2004-Nov. 2005; since 2006, bi-monthly staff visits).
  • Exchange Rate Issues: Foreign Exchange Risk Management, Dealing with Capital Inflows, Transition between Exchange Rate Regimes, Operational Preparations for Exchange Rate Flexibility (TA missions in April, July, and December 2005).
  • Public Debt Management Strategy and Debt Market Development, (TA missions in July 2005 and January 2006).
  • Setting Up an Action Plan for Transiting to Inflation Targeting (TA mission October 2005).
  • Operational Advice on Improving Monetary Policy Operations (TA mission February 2006).
  • Monetary Policy Communication Strategy (TA mission March 2006).
  • Strengthening Banking Supervision (Resident advisor to the NBU, April-October 2005); Instruments and Procedures to Deal with Problem Banks (TA mission April 2005).

In addition to TA, the IMF has prepared a number of documents for the authorities, including on Ukraine’s competitiveness, pros and cons of various exchange-rate regimes and policy requirements, aspects for determining the optimal level of reserves, and issues in determining the weights of currency baskets.

In addition to TA, the IMF has prepared a number of documents for the authorities, including on Ukraine’s competitiveness, pros and cons of various exchange-rate regimes and policy requirements, aspects for determining the optimal level of reserves, and issues in determining the weights of currency baskets.

  • Monetary operations: The NBU already has a complete toolkit of market-oriented monetary policy instruments, including liquidity-absorbing and liquidity-injecting open market operations (Box II.4). But in the past, it has chosen to rely mainly on foreign-exchange interventions and reserve requirements. When transiting to inflation targeting, a short-term policy interest rate should take over from the exchange rate as the operating target. Preparations to set a policy rate, which will be steered in a corridor with standing facilities marking the upper and lower boundaries, are underway.
  • Transmission mechanism: The lack of a yield curve and an active securities market is also viewed as a hindrance for the effective transmission of interest-rate impulses to bank rates and inflation. Empirical work confirms that the transmission mechanism in Ukraine is still weak (Petryk and Nikolaychuk, 2006; Mykhaylychenko and others, 2004). The findings support the view that the exchange rate (and monetary indicators) could continue to play some role in the transition period and stress the need for the government to pursue a public debt management strategy with the objective to develop domestic securities markets. But at the same time, the corporate securities market is already growing quickly, albeit from a low base, thus supporting the transmission mechanism. Moreover, much of the missing link in the past was caused by the lack of an active NBU interest-rate policy and is expected to strengthen as the NBU policy changes. Empirical studies have found the transmission mechanism to become stronger under more flexible exchange rates and financial market development (Ganev and others, 2002; Tiemann, 2004).
  • Institutional changes: Institutional changes can help support the creation of a new transparent nominal anchor under an inflation targeting regime. These include legal changes that would provide the NBU with a clearer mandate to pursue the objective of price stability. Efforts to increase the NBU’s communication with the public, including by issuing an inflation report, are already under way.
  • The inflation target itself: Against the backdrop of expected large shocks to energy prices (in particular resulting from adjustments of gas import prices, but also due to oil price developments), targeting a core inflation index that excludes energy could be advantageous. The NBU and the State Statistics Committee are preparing such indices. At the same time, a transparent and predictable government policy on administrative price changes is needed to avoid undermining the NBU’s credibility in case of inflation target misses that derive from specific administrative price adjustments.

Box II.4.The NBU’s Toolkit of Monetary Policy Instruments

The NBU has a wide range of market-oriented policy instruments in its arsenal, but has relied mainly on foreign-exchange interventions and changes in required reserves.

Foreign-exchange interventions: Under the de facto peg, the main liquidity impact has come from NBU interventions in the spot foreign-exchange market. Currently, the NBU intervenes to keep the hrvnia in a band of Hrv/US$5.00–5.06.

Reserve requirements: Banks need to hold unremunerated required reserves for all demand, time, and savings deposits. The reserve ratios are differentiated by maturity and currency denomination and relatively low at an effective rate of 2.2 percent (down from 6.8 percent in April 2006). Currently, banks have to maintain the entire amount of required reserves on a daily basis; cash in vault is excluded from the reserve calculation. In the past, the NBU frequently changed all of the required reserve criteria to affect banks’ liquidity conditions.

Standing facilities: The NBU operates two marginal overnight lending facilities: one against collateral and one without collateral. The latter is priced at 100 basis points above the fomer. There are several restrictions on the amount and frequency of bank borrowings, some of which are linked to banks’ credit quality. Banks use uncollateralized overnight loans frequently, often toward the end of the month to fulfill reserve requirements. The NBU also has deposit facilities at its disposal, at various maturities (2–7 days, 8–21 days, and 22–30 days), but due to the low interest rates banks have rarely used them (Box table).

Money market instruments: The NBU can inject liquidity through refinancing loans for up to one year, repurchase arrangements, and outright operations. None of the operations has been frequently used and tenders are typically announced ad hoc. While the NBU also sets a discount rate, no transactions are conducted at that rate. To absorb liquidity, the NBU can offer deposits and NBU certificates of deposits, with maturities up to one year. In 2004–05, the NBU issued CDs quite frequently, though the volumes were usually quite low (the largest stock outstanding was Hrv4.8 billion or 7.5 percent of base money in June 2005). NBU CDs were auctioned and maturities ranged from 1–270 days, on average about 60 days.

Ukraine: NBU Monetary Policy Operations, Jan. 2005-Nov. 2006
Interest rateAmountNumber of banks
200520062005200620052006
Standing facilities(Percent)(Millions of hryvnias)
Overnight loans (collateralized)12.010.73941,10485
Overnight loans (uncollateralized)15.012.710,7574,4284440
Deposit facility4.51.01576,13538
Money market operations
NBU loans12.910.92328881221
Repurchase agreements12.010.86131,30501
NBU CDs3.84.714,78860272
Reverse repurchase agreements1.00.02,300010
Sources: National Bank of Ukraine; and staff estimates.
Sources: National Bank of Ukraine; and staff estimates.
Appendix II.1. External Sector Data Issues

Trade Data Issues

There have been several possible discrepancies in recent Ukrainian trade data. During 2004, customs data for goods imports fell some $4½ billion (or 15 percent) short of the corresponding data based on banking sources. For surging exports, concerns have been expressed about over-reporting. During 2005, exports were much weaker than projected, declining in volume terms by 8½ percent, far short of model-based forecasts (5 percent).

Comparing Ukrainian data with the IMF’s Direction of Trade (DOT) Statistics suggests that under-reporting may have been an issue in some instances (Table II.A.1). For 2004 imports, the DOT data exceed that reported by Ukrainian customs (but fall well short of the banking data). For 2004 exports, the DOT data do show a substantial jump, and, if anything, exports may have been under-reported. Export under-reporting could also be an issue in 2005: if the DOT number is the “true” export number, export volume growth would be much closer, but still somewhat below the model-based projection.

Table II.A.1.Ukraine: Trade Data
20042005
(In $ millions)
Imports (customs data)29,69136,159
Imports (DOT data)31,27439,929
Discrepancy (in percent)-5.1-9.4
Exports (Ukrainian data)32,62933,959
Exports (DOT data)34,23836,628
Discrepancy (in percent)-4.9-7.9
Sources: IMF Direction of Trade statistics and National Bank of Ukraine
Sources: IMF Direction of Trade statistics and National Bank of Ukraine

A disaggregated look at exports in 2005 reveals specific areas of concern. For machinery and equipment, the decline was far too steep to be explained on competitiveness grounds alone. This is an area where export under-reporting may have been most prevalent. For metal exports (predominantly steel), the value/volume decomposition is open to question: whereas official data imply that Ukraine’s steel export prices (in dollars), increased by a robust 17 percent, commercially available indices such as Bloomberg point to a decline.

Tax evasion and capital flight probably underlie trade data discrepancies. Under-invoicing of imports to the customs authorities may relate to purely domestic transactions that are reported as imports to avoid payment of VAT. Over-reporting of exports may also reflect tax motivations. However, the bulk of the discrepancy between bank and customs import data in 2004 is considered to reflect capital flight—and was accordingly classified under short-term capital flows in the balance of payments.

Improving trade data is clearly an issue that needs to be addressed in the context of improvements of Ukraine’s national accounts statistics.

International Investment Position (IIP) Data

There has been a persistent discrepancy between the reported changes in stocks of external assets and liabilities on the one hand, and the corresponding balance of payments flows on the other. From a quantitative perspective, the discrepancy is very large: the net cumulative short-term capital outflow between 2000 and 2005 amounted to $18.2 billion, against a recorded overall net external liability position of $28.8 billion at end-2005.

In Ukraine, short-term capital flows (which in the past included some transactions mis-classified under equity investment) are not taken into account in compiling the IIP.

Scant information on these transactions does not allow their classification into appropriate IIP categories. Progress in this area is important to get a better sense of the types of capital movements involved, and possibly shed some light on whether there is also an issue of mis-classification of current transactions that could account for some of the above trade puzzles.

Appendix II.2. Macroeconomic Balance Approach: Variable Definitions and Data Sources

Dependent variable: Current account as a ratio to GDP (source: IMF World Economic Outlook).

Explanatory factorVariable definitionData source
Fiscal balanceGeneral government balance as a ratio to GDPWEO
Foreign direct investmentNet FDI as a ratio to GDPWEO
Relative incomePPP-based per capita income as a share of the U.S. levelWEO
DemographicsShare of 65 and older in the population; population growth rateWorld Bank, WDI
Commodity terms of tradeEnergy balance as a ratio to GDPWEO and country desks
Reserve coverageOfficial reserves in months of imports of goods and servicesWEO
References

    CheungY-W.M.Chinn and E.Fujii2006The Overvaluation of Renminbi UndervaluationManuscript, University of California Santa Cruz.

    • Search Google Scholar
    • Export Citation

    ChinnM. and E.Prasad2003Medium-Term Determinants of Current Accounts in Industrial and Developing Countries: An Empirical ExplorationJournal of International Economics Vol. 59 no. 1.

    • Search Google Scholar
    • Export Citation

    GanevG.K.MolnarK.Rybinski and P.Wozniak2002Transmission Mechanisms of Monetary Policy in Central and Eastern EuropeCASE Report No. 52.

    • Search Google Scholar
    • Export Citation

    HigginsM.1998Demography, National Savings, and International Capital FlowsInternational Economic Review Vol. 39 no. 2.

    HoC. and R.N.McCauley2003Living with Flexible Exchange Rates: Issues and Recent Experiences in Inflation Targeting Emerging Market EconomiesBIS Working Paper No. 130.

    • Search Google Scholar
    • Export Citation

    HusainA.M. (2006)To Peg or Not to Peg: A Template for Assessing the NoblerIMF Working Paper 06/54 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    International Monetary Fund2005World Economic Outlook September 2005. Chapter IV.Does Inflation Targeting Work in Emerging Markets? pp. 161-186.

    • Search Google Scholar
    • Export Citation

    IsardP. and H.Faruqee1998Exchange Rate Assessment: Extensions of the Macro-economic Balance ApproachIMF Occasional Paper No. 167.

    • Search Google Scholar
    • Export Citation

    IsardP.H.FaruqeeR.Kincaid and M.Fetherston2001Methodology for Current Account and Exchange Rate AssessmentIMF Occasional Paper No. 209.

    • Search Google Scholar
    • Export Citation

    KrajnyákK. and J.Zettelmeyer1998Competitiveness in Transition Economies: What Scope for Real AppreciationIMF Staff Papers Vol. 45.

    • Search Google Scholar
    • Export Citation

    LaneP. and G.M.Milesi-Ferretti2002External Wealth, the Trade Balance, and the Real Exchange RateEuropean Economic ReviewJune.

    • Search Google Scholar
    • Export Citation

    LeheydaN.2005Determinants of Inflation in Ukraine: A Cointegration Approach” (mimeo).

    LissovolikB.2003Determinants of Inflation in a Transition Economy: The Case of UkraineIMF Working Paper 03/126 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    MykhaylychenkoM.V.Hryniv and W.Paczynski2004Monetary Transmission in Ukraine: a VAR Approachpaper presented at the conference on “Sustaining Low Inflation in Ukraineorganized by the Center for Social and Economic Research.

    • Search Google Scholar
    • Export Citation

    ObstfeldM. and K.Rogoff1996Foundations of International MacroeconomicsMIT Press, Cambridge, MA.

    PetrykO. and S.Nikolaychuk2006Structural Model of the Transmission Mechanism of the Monetary Policy in UkraineNational Bank of Ukraine.

    • Search Google Scholar
    • Export Citation

    RogerS. and M. R.Stone2005On Target? The International Experience with Achieving Inflation TargetsIMF Working Paper 05/163.

    • Search Google Scholar
    • Export Citation

    SchiffJ.P.Egoumé-BossogoM.IharaT.Konuki and K.Krajnyàk2006Labor Market Performance in Transition. The Experience of Central and Eastern European Countries,IMF Occasional Paper 248.

    • Search Google Scholar
    • Export Citation

    SiliverstovsB. and O.Bilan2005Modeling Inflation Dynamics in Transition Economiesin: Eastern European EconomicsVo. 43 no. 6pp.66-81.

    • Search Google Scholar
    • Export Citation

    TiemannAlexander2004Interest Rate Pass-Through in Romania and Other Central European CountriesIMF Working Paper 03/211.

    TiffinA.2006Competitiveness, Convergence, and the Equilibrium Real Exchange Ratein: IMF Country Report No. 05/20pp.36-46 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
24See Ukraine: Ex Post Assessment of Long-term Use of Fund Resources (2005), Box 9, for a sketch of the debate on the exchange rate regime between the Ukrainian authorities and the IMF since 1995.
25Prepared by Ioannis Halikias and Mark Flanagan.
27The energy balance (as a share of GDP) is used rather than energy prices to help distinguish the impact on the current account between net energy importers and exporters.
28To avoid causality problems, lagged reserve coverage is used for estimation purposes. Net foreign assets can best capture solvency considerations and have been widely used in empirical work—see Lane and Milesi-Ferretti (2002) and Chinn and Prasad (2003). Data limitations—explained in Appendix I—preclude the use of this variable here.
29Other variables were dropped from the final specification due to low explanatory power (low statistical or economic significance).
30These are long-run elasticities. They need to be treated with caution since (i) the sample period over which they are estimated is short; (ii) there have been large recent changes in Ukraine’s geographic and commodity structure of trade; and (iii) there may be non-linear responses in some industries, reflecting rigid production structures.
31See Cheung, Chinn and Fujii (2006) for a recent application of the approach to China.
32See Tiffin (2004), which in turn builds on previous cross-country work on transition economies in Krajnyak and Zettelmeyer (1998).
33Prepared by Andrea Schaechter.
34The IMF Annual Report on Exchange Arrangements and Exchange Restrictions provides the classification of exchange rate regimes.
35When active NBU policies to change monetary conditions occurred, they aimed mainly on loosening liquidity conditions, for example through lower reserve requirements. Active sterilization operations in times of large excess liquidity happened in 2005, but they were limited in volume and interest rates remained low.
36For a comparison of labor market policies institutions in Central and Eastern European Countries, see Schiff and others (2006)
37See Chapter I for more details on the macroeconomic effects of the energy price shock.
38See Leheyda (2005), Lissovolik (2003), and Siliverstovs and Bilan (2005).
39The flow-of-funds analysis in Chapter III.B indicates that corporates may have built up external financial asset partly financed by loans from their own banks. These banks, in turn, obtained part of their funding from abroad, including from offshore businesses of the same corporate and financial groups.

Other Resources Citing This Publication