Chapter

3 The International Community and the CIS-7 Countries

Author(s):
Sarosh Sattar, and Clinton Shiells
Published Date:
April 2004
Share
  • ShareShare
Show Summary Details
Author(s)
Philip R. Lane*

Overview

Without doubt, the collapse of the communist bloc and the dissolution of the Soviet Union during 1989–1991 represented the largest regime change experienced in the world since the 1940s. In terms of economic policy, the countries that emerged from the Soviet bloc faced major challenges in terms of re-molding institutions and markets to deliver growth and prosperity for their citizens. The scale of the adjustment problem was most acute for the countries in the former Soviet Union. These countries had economic structures that were directed toward fulfilling specialized roles within the Soviet central planning system: for this group, the challenge of building self-functioning market-based economies was especially severe.

From the beginning, the international community recognized that assistance should be provided to help in the transition process in Central and Eastern Europe and the newly independent states of the former Soviet Union. Regarding the latter group, the Economic Declaration of the July 1992 G-7 Munich Summit promised1

“We offer the new States our help for their self help.”

This position recognized that the primary responsibility for development resided with the countries of the former Soviet Union but that their efforts should be supported and facilitated by international aid. The G-7 Munich declaration went on to state

“We welcome the membership of the new States in the IFIs. This will allow them to work out economic reform programs in collaboration with these institutions and on this basis to make use of their substantial financial resources. Disbursement of these funds should be linked to progress in implementing reforms.”

The international community thereby charged the international financial institutions (IFIs) with a central role in assisting the development efforts of these countries. In tandem with their new responsibilities in the other transition countries in Central and Eastern Europe, this represented a major new challenge for the multilateral organizations.

This statement also explicitly highlighted the importance of conditioning financial support to the country authorities’ successful restructuring of their economies: it was envisaged that aid would be allocated in a discriminating fashion, and the withdrawal of support was threatened for nonperformers. The text also indicates the desirability of a partnership model by which these countries would actively contribute to the design of the reform programs that would be approved by the IFIs.

The goal of this paper is to examine the contribution of the international community to the development efforts in the CIS-7 countries. The primary focus is on the role played by the IFIs that have been active in these countries—the IMF, the World Bank, the European Bank for Reconstruction and Development (EBRD), and the Asian Development Bank—in line with their predominant role as the main conduits for international assistance to these countries. We will also address the other mechanisms by which the international community influences the development of these countries.

The structure of the rest of the paper is as follows. In the second section, we briefly review the debate about appropriate strategy for transition countries. Some basic facts about the CIS-7 countries are presented and discussed in the third section. The fourth section describes and analyzes the involvement of the IFIs in the CIS-7 economies. We turn in the fifth section to addressing the future role of the international community in assisting the CIS-7 countries. Finally, some concluding remarks are offered in the last section.

The Transition Debate

From the beginning, there was little disagreement among analysts that the ultimate goal was to establish market-based systems in the transition countries. Moreover, it was also widely accepted that macroeconomic stabilization was a required component of any reform strategy.

Beyond those core elements, there has been considerable debate about the relative merits of “big bang” versus “gradualist” approaches to reform. Adherents to the former approach took the view that it was best to move rapidly in establishing a market-based system, liberalizing prices and trade as quickly as possible and moving enterprises out of state hands through small- and large-scale privatization programs. At a political economy level, it was argued that a rapid transition would prevent the emergence of blocking coalitions of antireformist groups that gained from the status quo or feared change.

Proponents of gradualism argued that it inevitably takes time to establish the institutions required to ensure that markets operate efficiently and to avoid excessive social disruption. Indeed, there is no unique set of institutions that is required to have a successful market economy. In this regard, the industrial countries vary quite significantly in terms of legal codes, welfare programs, and financial systems.

Moreover, gradualists claim that popular support for reform could actually be enhanced by a partial, gradualist approach: for instance, a targeted initial reform may assuage fears and in itself create new interest groups that will support and lobby for stages of reform.2,3 With respect to building institutional capacity, even big bang adherents recognized the importance of devoting resources to establishing appropriate legal codes and regulatory and tax systems.

At an intermediate level, it seems sensible to hold the position that the appropriate speed and nature of reform will differ across countries depending on their initial and economic structures.4 In contrast, countries that began reform as agricultural-oriented rural economies with limited social welfare systems, such as China in 1978, could perhaps afford to take a more gradualist approach. Even the reform of agriculture was easier in China than in the former Soviet Union, due to peasant farming in the former as compared to the collectivized farm system established under Soviet rule.

International and locational aspects also determine the nature of reform efforts in countries.5 In their framework, institutions vary across countries and must be discovered through policy experimentation. A country that is a close neighbor to a leader has the option to imitate that country’s institutions, and, being fundamentally similar, this is likely to be a successful strategy. At the other extreme, a remote country may not enjoy the demonstration effect provided by a neighboring leader but may achieve success through unfettered experimentation. Because remote countries experiment in order to identify appropriate institutional structures, it also follows that there is likely to be a high variance in outcomes among this group.

The worst off are the intermediate countries. These may be tempted to imitate a leader that is not too far away, but this may be a flawed strategy because their structural differences are such that attempts at institutional transplantation may fail. An additional prediction of this theory is that corruption is an increasing function of remoteness: governments in countries that are closer to leaders are more constrained in their actions.

Mukand and Rodrik conduct an empirical test using the growth and reform experiences of the post-socialist countries. They find a U-shaped relation between growth and distance from the European Union (EU), which is the natural leader for these countries. The fastest-growing transition countries have been those closest to the European Union, but the more remote countries—Kazakhstan, the Kyrgyz Republic, and Uzbekistan—have done better than intermediate cases, such as Azerbaijan, Georgia, Moldova, Russia, and the Ukraine. Although intriguing, the Mukand-Rodrik hypothesis only explains the dire performance of a remote country such as Tajikistan by attributing it to institutional experimentation that ex post proved to be a failure. Moreover, it is plausible that Russia, rather than the European Union, is the natural anchor for these countries: the instability of the Russian reform process during the 1990s failed to provide much by way of positive example for the CIS-7 countries.

In terms of application to the CIS-7 countries, there are conflicting elements in this debate. At one level, very rapid reform may have been especially difficult for these countries in view of the scale of the development challenges they faced and the absence of a natural policy anchor, such as the prospect of eventual EU membership provided for the Central and Eastern European transition countries. In related fashion, the initial conditions and political structures of these countries presented the IFIs with a set of challenges different from the more advanced and Western-orientated transition countries. At another level, the debate on the sequencing of reforms is quite relevant for these countries; for instance, price liberalization has been considerably easier than tackling more deep-seated institutional deficiencies.

With respect to the political economy of transition, a slow pace of stabilization and reform has risked leaving some of these countries in an under-reform trap by allowing the formation and consolidation of antireform lobbies and political groups.6 In addition, it is unclear the extent to which the formation of dynamically sustainable, pro-reform coalitions can be easily predicted in societies with complex ethnic and political structures: designing an optimal gradualist transition strategy for such a country would have been extremely forbidding.

The CIS-7: Basic Features

It is well understood that during the period following the breakup of the Soviet Union, there was a much deeper and more prolonged initial output decline and delayed stabilization for the CIS-7 countries than for most of the transition economies in Central and Eastern Europe. In this regard, an obvious question to ask is whether the CIS-7 countries were slower reformers than other transition economies. To address this question, Table 3.1 reports the results for pooled regressions of the form,

where the dependent variable is one of the reform indices (indexed by j) reported by the EBRD in its Annual Transition Report and CIS7 is a 0–1 dummy variable that takes the value 1 for CIS-7 countries and 0 for the other transition countries. The data period is 1991–99, and the values of the EBRD indicators range from 0 to 4, with a higher value indicating a higher level of reform. With the exception of the price liberalization index, the CIS7 dummy variable is significantly negative in all cases, confirming that these countries as a group have lagged behind the reform efforts in other transition countries.7

Table 3.1.The CIS-7 and Reform: 1991–99 Panel Estimation1
CIS-7AdjR2NPeriod
Price liberalization-0.160.012251991–99
(1.62)
Trade liberalization-0.860.092251991–99
(4.91)***
Small-scale privatization-0.690.072251991–99
(4.42)***
Large-scale privatization-0.350.022251991–99
(2.59)**
Enterprise reform-0.510.102251991–99
(6.05)***
Competition policy-0.490.022251991–99
(3.32)***
Infrastructure reform-0.560.12501998–99
(3.30)***
Banking sector reform-0.610.112221991–99
(6.15)***
Reform of nonbanking-0.570.142201991–99
(7.10)***
Legal extensiveness-0.300.02701998–99
(1.68)*
Legal effectiveness-0.650.11701998–99
(3.68)***
Source: EBRD Annual Transition Reports.

t-statistics in parentheses; ***, **, and * denote significance at the 1, 5, and 10 percent levels, respectively.

Source: EBRD Annual Transition Reports.

t-statistics in parentheses; ***, **, and * denote significance at the 1, 5, and 10 percent levels, respectively.

Table 3.2 repeats the exercise for 1999, the last year in our dataset. At least for some categories, there is evidence of catch-up: the CIS7 dummy variable is no longer significant for trade liberalization, large-scale privatization, and the legal reform indicators. Significant differences in the level of reform between the CIS-7 countries and others in the transition group remain, however, for the other categories.

Table 3.2.The CIS-7 and Reform: 1999 Cross Section1
CIS-7AdjR2N
(1) Price liberalization-0.05-0.0425
(.29)
(2) Trade liberalization-0.29-0.0325
(.60)
(3) Small-scale privatization-0.400.0325
(1.72)*
(4) Large-scale privatization-0.21-0.0325
(.74)
(5) Enterprise reform-0.460.1125
(2.98)***
(6) Competition policy-0.450.0925
(2.29)**
(7) Infrastructure reform-0.550.0825
(2.10)**
(8) Banking sector reform-0.660.1025
(2.47)***
(9) Reform of nonbanking-0.640.1625
(2.75)***
(10) Legal extensivenses-0.360.0125
(1.16)
(11) Legal effectiveness-0.530.0625
(1.54)
Source: EBRD Annual Transition Reports.

t-statistics in parentheses; ***, **, and * denote significance at the 1, 5, and 10 percent levels, respectively.

Source: EBRD Annual Transition Reports.

t-statistics in parentheses; ***, **, and * denote significance at the 1, 5, and 10 percent levels, respectively.

The CIS-7 countries were especially slow to reform in the initial 1991–94 period. In most cases, economic development and reform took second place to the struggle between rival blocs for political supremacy, spilling over into civil war in Tajikistan, a major insurgency in Georgia, the Transnistria conflict in Moldova, and the Armenia and Azerbaijan conflict over Nagorno-Karabakh. The comparative weakness of internal political systems in these countries put them at a substantial disadvantage compared to the majority of Central and Eastern European countries. In some cases, officials from the Russian minority within these countries were pushed out for political reasons, and their replacements lacked similar training or experience.

This brief review suggests that the CIS-7 countries as a group are indeed distinctive in terms of their transition experience. Although significant growth and reform have been achieved since the mid-1990s, these countries did not turn the corner as quickly as—and still lag largely behind—the more advanced transition countries in Central and Eastern Europe. In this context, it is important to address the question of how the international community, and the IFIs in particular, have performed in assisting the transition process in these countries and whether alternative strategies to those currently pursued can better tackle the considerable developmental challenges that still face these countries.

The IFIs and the CIS-7

As was noted in the introduction, the international community responded to the destruction of the communist bloc by charging the IFIs with the task of aiding the transition of the emerging postcommunist independent states to normal market economies. Indeed, the EBRD was established in direct response to this challenge as a complementary agency to the activities of the already-established multilateral institutions.

We can think of the IFIs as providing three kinds of assistance: policy advice, technical assistance, and financial support. With regard to the first category, and in line with its mandate, the IMF has taken the lead in providing advice on macroeconomic stabilization, balance of payments issues, and fiscal affairs. It has also advised in related areas (such as banking stability and banking regulation) and on broader items on the structural reform agenda.

In addition to advice on individual projects and sectoral issues, the World Bank has also been a lead advisor on structural adjustment, public finances, and financial sector reform. Policy surveillance and assessments are central to the regular Article IV missions conducted by the IMF and to the Country Assistance Strategy (CAS) reports prepared by the World Bank, in addition to influencing the conditions attached to the loan and assistance agreements that are negotiated with the individual countries. The policy advice offered by these institutions is now in principle integrated into the Poverty Reduction Strategy Paper (PRSP) process, as discussed later.

At one level there is a broad consensus about the correct ultimate policy targets for these countries. In addition to macroeconomic stability, obvious elements of reform include price and trade liberalization, privatization of small-and large-scale enterprises, the provision of basic infrastructure, the establishment of an investment-friendly business environment, and the implementation of appropriate legal codes and appropriate regulatory structures. Of course, even with a general acceptance of the desirability of all these elements of reform, this still leaves plenty of room for debate about appropriate sequencing and the relative priorities among this wish list. There also remains the core issue for the IFIs of the appropriate financing of national development strategies. In addition, the coordination of policy advice to ensure a consistent and streamlined framework presents a considerable management challenge for the multilateral agencies.

Evaluating the quality of policy advice from the IFIs, however, is extremely problematic. Sovereign nations are free to disregard or implement only loosely external policy advice, such that economic outcomes cannot be closely linked to the policy advice given. Of course, economic performance is also influenced by various shocks that may disrupt the forecasts generated by policy analysts. In addition, country reports by the IFIs do not reveal the full degree of interaction between staff members and the national authorities in policy formation: policy errors may be avoided by private and informal interventions, and published documents tend to focus only on a negotiated list of specific policy recommendations rather than on the entire dynamic process of interaction between external advisors and domestic policymakers.

Turning to the policy record of the CIS-7 countries, it seems apparent that the major problem so far has been a failure to implement reforms sufficiently rather than the adoption of possibly erroneous policies that were recommended by the external agencies. A more serious concern relates to advice concerning financial policy, in view of the high debt burdens in some of these countries: we turn to this issue below.

At a more challenging level, it could also be argued that the IFIs might have done more to understand the political obstacles to rapid reform in the CIS-7 countries and to assist in the design of appropriately gradualist strategies that build a reform momentum cumulatively. Aside from the difficulty of correctly selecting such strategies in real time and without the benefit of hindsight, the fundamental issue is was it appropriate for the IFIs to become too involved in domestic political decisions.

In terms of technical assistance, these institutions devoted considerable resources providing technical advice to the CIS-7 and other transition countries.8 This took various forms: from short-term training courses to financing long-stay technical experts to advise on the establishment of central banks, budgetary procedures, corporate and bankruptcy law, and financial regulation.

The delivery of such technical assistance presented the multilaterals with substantial resource and logistical demands. The success of these programs was inevitably greater in the more straightforward areas, such as the establishment of a functioning central bank, than in achieving rapid progress in improving the quality of governance in softer areas, such as establishing a predictable rule of law. It also relied on domestic governments to exploit the lessons from short-term training modules and to use long-stay resident advisors efficiently. To a significant degree, the flow of trained personnel from the public to the private sector, especially in the financial sector area, meant that much technical assistance required periodic repetition to train new cohorts of policy officials. This cannot, however, really be interpreted as being of zero value in that there is a social benefit to building up the expertise of the private sector in these countries.

Table 3.3 reports aggregate aid flows to the CIS-7 countries over 1995–2000. Overall, aid flows have represented a considerable portion of national incomes. There has also been substantial cross-country variation, however: average aid flows ranged from only 1 percent of GDP for Uzbekistan to 16.9 percent for the Kyrgyz Republic. This variation is not correlated with output per capita (the correlation is an insignificant -0.06) or with the extensiveness of reform (the correlation with a composite of the EBRD indicators is only -0.12).

Table 3.3.Total Aid Flows to the CIS-7(As a percentage of gross national income)
199519961997199819992000Average
Armenia7.618.310.17.511.311.211.0
Azerbaijan4.93.14.72.33.72.83.6
Georgia11.510.36.54.68.45.67.8
Kyrgyz Republic19.512.914.114.424.116.716.9
Moldova2.22.13.32.38.99.04.6
Tajikistan2.910.49.68.411.915.29.8
Uzbekistan0.80.61.01.10.91.41.0
Source: World Bank, World Development Indicators online database.
Source: World Bank, World Development Indicators online database.

Table 3.4 reports some broad external debt statistics for the CIS-7 countries for the years 1995 and 2000. We take 1995 as a useful initial reference point because lending to most of the CIS-7 countries was more restricted before 1995 due to political instability and internal military conflicts. Moreover, there was an initial bilateral learning period, as the IFIs studied these little-known economies and the governments of these countries gained awareness of the financing options that were available from the multilaterals.9 The failure to achieve early macroeconomic stabilization also disqualified these countries from significant levels of funding during the initial years following independence.

Table 3.4.External Debt Statistics
Debt/GNI1LT/Total2MLT3BILAT3BANK3
1995200019952000199520001995200019952000
Armenia134680767072302603
Azerbaijan1124645818465226027
Georgia6854848024376760103
Kyrgyz19150788339486131021
Republic
Moldova2391668547394712649
Tajikistan28125938501788141242
Uzbekistan7991171255402849
Source: World Bank, Global Development Finance online database.

Debt/GNI is the ratio of total external debt to gross national income.

LT/total is the ratio of long-term to total external debt.

MLT, BILAT, and BANK refer to the shares of long-term debt that are owed to the multilateral institutions, bilateral donors, and private lenders, respectively.

Source: World Bank, Global Development Finance online database.

Debt/GNI is the ratio of total external debt to gross national income.

LT/total is the ratio of long-term to total external debt.

MLT, BILAT, and BANK refer to the shares of long-term debt that are owed to the multilateral institutions, bilateral donors, and private lenders, respectively.

Since 1995, the CIS-7 countries have experienced sharp increases in the ratio of external debt to output; debt growth was most dramatic in the Kyrgyz Republic, Moldova, and Tajikistan. The fallout from the ruble devaluation by Russia in 1998 also contributed to the adverse debt dynamics because many of these countries responded with large depreciations vis-à-vis the U.S. dollar that had the effect of increasing the real burden of foreign currency debt liabilities.

Excessive debt levels are now a concern in several of these countries. Georgia and the Kyrgyz Republic have both negotiated Paris Club agreements (in 2001 and 2002, respectively) that have led to some reduction and rescheduling of bilateral debts, and others (e.g., Armenia and Moldova) have also renegotiated debts with some individual creditors.10

Apart from oil-rich Azerbaijan, debt liabilities are overwhelmingly long term in maturity. As is clear from Table 3.4, however, this does not reflect a capacity to issue long-term debt to private investors but rather a heavy reliance on official financing. In some countries (Moldova, Tajikistan, and Uzbekistan), long-term debt owed to private institutions has increased noticeably. The large rise in official external debt relative to output during this period does not suggest an unwillingness by the IFIs to provide financing for these countries. Indeed, it may even represent overlending to the extent that borrowed funds were inefficiently deployed and the accumulated debt burden is constraining the development strategy in some of these countries. In turn, overlending may be the result of excessive optimism on the part of the multilaterals in terms of the anticipated growth performance of these economies, or it may reflect an inability to deprive badly behaved governments of official funding. In the latter case, international resources are wasted, and debtor moral hazard is encouraged, in that failure to reform need not involve sustained suspension from borrowing. The IFIs may, in the short term, raise the specter of nonagreement to a new financial program or to disbursement of certain funds, but the pressure to stay engaged in such countries means that it is much more difficult to withdraw from lending for a longer period of time.

The terms of multilateral financing are an important topic for debate. There have been several prominent calls for a shift toward concessional loan terms and outright grants in financing low-income reformist countries. In this regard, there is considerable variation among the CIS-7 countries, as shown in Table 3.5. At one extreme, the poorest country, Tajikistan, has enjoyed concessional terms on almost all of its multilateral debt. At the other extreme, the multilateral debt of Uzbekistan has been incurred almost entirely at nonconcessional terms. This mix has also changed over time. For instance, World Bank lending has increasingly taken the form of concessional IDA loans. The Poverty Reduction and Growth Facility (PRGF) program at the IMF has also improved the generosity of the terms of its lending to these countries.

Table 3.5.Importance of Concessional Flows from International Financial Institutions, 1994–20001(In percent)
Regional
IMFWorld BankDevelopment Banks
Armenia7998
Azerbaijan100
Georgia78100
Kyrgyz Republic10078
Moldova734
Tajikistan77100100
Uzbekistan1
Source: World Bank, World Development Indicators online database.

Ratio of concessional to total cumulative net financial flows.

Source: World Bank, World Development Indicators online database.

Ratio of concessional to total cumulative net financial flows.

Having described the financial support provided by the IFIs to the CIS-7 countries, we turn now to asking some questions about the effectiveness of multilateral lending to this pool of countries. In Table 3.6, we examine whether an increase in multilateral debt is associated with an increase or decrease in other sources of financing, such as bilateral debt and private debt. This is an important issue because it is essential to know the joint behavior of multilateral and other sources of funds in order to understand the total net impact on the recipient country. We have run regressions of the form

where X denotes bilateral or private per capita debt and MLT denotes multilateral debt. We consider two variants: pooled and fixed-effects specifications. From rows (1) and (2) of Table 3.6, we observe a positive relation between the levels of multilateral debt and bilateral debt. We do not claim to show that this is a causal relationship but merely that these variables are significantly correlated. A causality test would require much more information than is available. Regarding private debt, the pooled specification in row (3) does not show a significant relation with multilateral debt, but this is overturned in the fixed-effects specification in row (4). It is interesting to note that rows (2) and (4) yield the same point estimates: the within-country comovement with respect to multilateral debt is the same for debt from private lenders as debt from bilateral creditors.

Table 3.6.Credit Multiplier and Multilateral Lending1
MultilateralAdjR2N
Bilateral(1)0.160.0641
(3.71)***
(2)0.210.9741
(9.44)***
Bank(3)0.02-0.0242
(0.34)
(4)0.210.3342
(3.23)***
Source: Joint BIS-IMF-OECD-World Bank external debt database.

t-statistics in parentheses; *** denotes significance at the 1 percent level. Rows (1) and (3) represent a pooled specification, while rows (2) and (4) represent a fixed-effects specification.

Source: Joint BIS-IMF-OECD-World Bank external debt database.

t-statistics in parentheses; *** denotes significance at the 1 percent level. Rows (1) and (3) represent a pooled specification, while rows (2) and (4) represent a fixed-effects specification.

The fixed-effects results in Table 3.6 indicate that a credit multiplier does exist for the CIS-7 countries in the sense that multilateral debt appears to leverage bilateral and private lending. To some extent, this may be the deliberate result of cofinancing arrangements (e.g., between the EBRD and some bilateral donors). More generally, it reflects possibly the important role played by multilateral lending in signalling the quality of the investment climate in a given host country: if a country fails to qualify for multilateral financial assistance, it may be viewed as a poor candidate for bilateral or private funding. A more pessimistic interpretation runs along the overborrowing lines: in political settings where multiple powerful groups have access to fiscal resources, an increase in official financial assistance may prompt a disproportionate increase in inappropriate expenditures.11

The analyses in Tables 3.73.9 ask whether debt flows respond to reforms in the CIS-7 countries. The answer to this question may provide some information for the conditionality debate. The standard prescription is that well-behaved reformers should be rewarded with increased financial support, whereas funding cannot be justified for serial nonreformers. Indeed, nonreformers may benefit indirectly from exclusion from funding if it stimulates reform in these countries. Against this view, if the multilateral institutions play the role of creditors-of-last-resort for the international community, humanitarian considerations may militate against the denial of funding to a desperately poor country, regardless of its reform record. Moreover, it is reasonable to assume that, at least in some finely balanced political equilibria, a persistent lack of financing may diminish the probability of reform by increasing the influence of isolationalists vis-à-vis outwardly orientated integrationist political blocs.12 Finally, another relevant consideration is that a country with a good reform record may be able to achieve access to private sources of credit and hence graduate from dependence on official financial flows.

Table 3.7.Multilateral Debt Flows and Reward for Reform1
MultilateralAdjR2NFE
Price liberalization(1)16.310.0835No
(5.41)***
(2)1.500.4735Yes
(1.36)
Trade liberalization(3)3.440.0535No
(1.99)**
(4)-9.350.6335Yes
(3.74)***
SINDEX(5)3.41-0.0335No
(.78)
(6)-21.300.5935Yes
(3.27)***
MINDEX(7)0.97-0.0235No
(.21)
(8)-21.100.6035Yes
(2.95)***
Sources: Debt data from joint BIS-IMF-OECD-World Bank external debt database; reform indicators from EBRD.

t-statistics in parentheses; *** and ** denote significance at the 1 and 5 percent levels, respectively.

Sources: Debt data from joint BIS-IMF-OECD-World Bank external debt database; reform indicators from EBRD.

t-statistics in parentheses; *** and ** denote significance at the 1 and 5 percent levels, respectively.

Table 3.8.Debt Flows and Reward for Reform: Bilateral Flows1
BilateralAdjR2NFE
Price liberalization(1)-1.030.0833No
(1.40)
(2)-1.120.4733Yes
(1.57)
Trade liberalization(3)0.370.0533No
(.51)
(4)0.170.6333Yes
(.25)
SINDEX(5)2.680.0833No
(3.80)**
(6)0.430.3533Yes
(.25)
MINDEX(7)2.740.0933No
(1.75)*
(8)0.350.3533Yes
(.19)
Sources: Debt data from joint BIS-IMF-OECD-World Bank external debt database; reform indicators from EBRD.

t-statistics in parentheses; ** and * denote significance at the 5 and 10 percent levels, respectively.

Sources: Debt data from joint BIS-IMF-OECD-World Bank external debt database; reform indicators from EBRD.

t-statistics in parentheses; ** and * denote significance at the 5 and 10 percent levels, respectively.

Table 3.9.Debt Flows and Reward for Reform: Bank Flows1
BankAdjR2NFE
Price liberalization(1)0.73-0.0335No
(.08)
(2)-7.60.5135Yes
(1.0)
Trade liberalization(3)6.110.0835No
(2.14)**
(4)10.950.5935Yes
(2.23)***
SINDEX(5)8.720.0135No
(1.48)
(6)23.950.5735Yes
(2.1)**
MINDEX(7)5.34-0.0235No
(1.03)
(8)24.430.5735Yes
(2.18)**
Sources: Debt data from joint BIS-IMF-OECD-World Bank external debt database; reform indicators from EBRD.

t-statistics in parentheses; *** and ** denote significance at the 1 and 5 percent levels, respectively.

Sources: Debt data from joint BIS-IMF-OECD-World Bank external debt database; reform indicators from EBRD.

t-statistics in parentheses; *** and ** denote significance at the 1 and 5 percent levels, respectively.

Our regression specification is

where FLOWitX is the year-t per capita financial flow from lender class X (multilaterals, bilaterals, and private banks) and REFORMit-1 is the value of a given reform indicator from year t-1. In order to conserve space, we do not report results for each individual EBRD reform indicator. Rather, we just include the two most important economic reforms (price and trade liberalization) and two composite indicators (SINDEX and M INDEX). SINDEX is just the unweighted average of the EBRD indicators that were earlier reported in Tables 3.1 and 3.2. We also omit the legal reform indicators because these are only available for 1998 and 1999. We also include the geometric average of the variables (MINDEX), to allow for possible among the various components of structural reform.13 The time period is 1996–2000.

The results for the pooled specifications in Table 3.7 do indeed show a positive cross-country relation between the level of reform and financial flows from the multilateral institutions: those countries that have done more to liberalize prices and trade receive more multilateral financial flows. The pooled results also confirm this for the key price liberalization and trade liberalization indices, such that these results hold across countries also. The pooled results are not significant only for the broader composite indices SINDEX and MINDEX, however: these variables are significant only along the time series dimension for multilateral flows. With the exception of the price liberalization variable, however, the fixed-effects specifications show that the within-country relation is actually negative: a country that improves its reform record experiences a decline in multilateral flows. As noted above, interpreting such a negative pattern is difficult. On the one hand, it may be consistent with the multilaterais’ failure to reward reform or, conversely, countries avoiding reform by relying on the availability of external financing. On the other hand, it may simply signal that reform improves access to private sources of financing.

The evidence for the relation between reform and bilateral financial flows is given in Table 3.8. In this case, we see no time series connection between reform and bilateral flows: a country that improves its reform performance does not experience greater bilateral financial inflows. At least for the composite indicators SINDEX and MINDEX, however, the pooled specifications do suggest that, in the cross section, those countries that are further along the reform path receive more bilateral support.

Finally, we turn to private debt flows in Table 3.9. The simple pooled specification does not do a very good job in explaining the cross-sectional variation in private debt flows: the results are typically insignificant, with the only exception being a positive relation between trade liberalization and private flows. The time series fixed-effects results are quite striking, however: if a country improves its reform record, it experiences a growth in private bank financial inflows. This result holds for trade liberalization, SINDEX, and MINDEX. The coefficient estimate for the price liberalization indicator is not significant. The lack of explanatory power for price liberalization throughout Tables 3.73.9 may just reflect that the major phase of price liberalization was largely completed before our sample period begins.

In combination with the fixed-effects results for multilateral flows in Table 3.7, the picture that emerges is reform tends to alter the composition of the source of financial flows away from multilaterals and toward private flows. Indeed, such a reconfiguration could be a desirable evolution as countries move up the reform ladder. To the extent that it does indeed reflect a failure of the multilaterals to sufficiently reward reform, however, it is potentially a cause for concern. Of course, the results presented here are only suggestive; a more comprehensive empirical framework would be required to test these hypotheses fully.

In summary, this section has reviewed some elements of the involvement of IFIs in the post-Soviet experiences of the CIS-7 countries. The institutions have certainly responded to the mandate of the international community to play an active role in the reform efforts of these countries on all fronts (policy advice, technical assistance, and financial support). Although errors have undoubtedly occurred—we will return to the high official debt burdens in some of these countries that are a matter for serious concern—it would be difficult to claim that the CIS-7 countries might have performed better without the intervention of these agencies. Nonetheless, the CIS-7 countries remain underdeveloped, as do many other low-income countries, and so it is appropriate to ask more from the IFIs in terms of finding new ways to promote development and growth.

A New Role for the International Community in Assisting the CIS-7

The past several years have seen much effort to reform the IFIs. Among the major influences on this debate have been the regular reviews at the G-7 summits. In addition, the U.S. Congress established an International Financial Institution Advisory Commission (the “Meltzer Commission”) that reported in 1999. The institutions themselves have also responded with self-critical reports and the initiation of major changes in their organizational and operational structures.14

Although the reform agenda was inspired in part by the large-scale international financial crises that were experienced during the 1990s, there was also considerable dissatisfaction with the performance of the IFIs in assisting the development low-income countries. For instance, a 1998 World Bank15 report provided a severe critique of the failure of aid to improve performance in countries that maintained distorted policy regimes, whereas it confirmed that aid did indeed play a useful role in countries with sound management that were firmly on the reform path.

A central principle in this new thinking is that national development strategies are most effective if “owned” by the domestic society. Ownership is established by ensuring the participation of the broad political establishment, the social partners, and civil society in the process of designing and evaluating development strategies and targets. The role of the international community in this process is to provide external advice and technical support as needed and to ensure that the design of financial assistance programs is consistent with the domestic development strategy. Consistency involves both arranging the financing required to make the strategy feasible and setting conditions on aid that reinforce the incentives to fulfill the targets specified in the strategy document.

At one level, this is the very philosophy underlying the Poverty Reduction Strategy Paper (PRSP) process that has been adopted by the IMF and World Bank. In turn, the PRSP feeds into an overall Comprehensive Development Framework (CDF) process that attempts to marshall the advice and assistance coming from the World Bank’s Country Assistance Strategy (CAS), the IMF’s Poverty Reduction and Growth Facility (PRGF), and the policy frameworks of the other multilateral development banks and the United Nations.

In this way, it is hoped that the assistance provided by the international community can be better streamlined and coordinated across the various agencies.16 Moreover, a comprehensive framework may also improve efficiency through facilitating greater specialization by individual institutions. Emphasizing the goal of poverty reduction in low-income countries is also intended to sharpen the focus of the multilateral development banks and the work of the IMF in these countries.

Although it is difficult to argue against a holistic approach to understanding the development needs of each country, there is concern that an excessively broad agenda may lead to mission creep and a loss of focus for individual agencies. In particular, the involvement of the IMF in long-term poverty reduction programs is troubling to some, it a departure from its traditional mandate.17 The counter-argument, however, is that overall consistency requires that all institutions be involved in understanding the broad development strategy for each individual country. This is perfectly consistent with the need for the lending and policy surveillance activities of each agency to remain specialized. The IMF focuses largely on advising and monitoring performance in terms of the appropriate macroeconomic policies while the multilateral development banks are more involved in the sectoral and institutional elements essential in achieving progress on reform and structural change.

In terms of development priorities, it is also recognized that the IFIs need to insist on and support better governance in borrower countries. In addition to providing financial and technical assistance to improve the quality of public sector management, develop expenditure programs, and implement legal and regulatory codes and practices, this also involves regular monitoring of the performance of borrower countries in improving governance. By implication, a country that persists with inadequate governance structures and antidevelopmental policies may be deemed a poor candidate for continued financial support. Moreover, under such circumstances, financial assistance may do more harm than good if it is allocated to inappropriate projects or disappears through corruption.

Striking the correct balance on this issue poses a major challenge for the IFIs. On the one side, the withdrawal of support from a regime that fails to promote development must be seen as a credible threat: there have to be circumstances under which this option is exercised. Given the poor reform record of the CIS-7 countries and the observed accumulation of high levels of external debt, the existing evidence suggests that this option could have been exercised at an earlier point in some of these countries. Although some factors behind the increase in the debt burden may not have been easy to forecast (e.g., the spillover from the 1998 Russian ruble devaluation and debt default), it is arguable that the multilaterals were too willing to lend to countries that had a small, short-term likelihood of engaging in reform. Of course, this kind of lending pattern has also occurred in other developing countries.18

It is quite understandable that the multilaterals may choose to remain engaged with countries that persistently fail to deliver. As previously noted, there may be a fear that nonengagement will strengthen the political hand of antireform political elements. In addition, the leaders of the IFIs may feel under pressure from the official policies of their major shareholders to be active in supporting these countries; the political fallout of walking away could be considerable.

There is no easy solution to this problem. Based on the history of the relations between the multilaterals and the CIS-7 countries, however, an essential element of a new start is to establish that a withdrawal of support will indeed occur in the event of an egregious failure to fulfill commitments. Combined with this tough love in response to nonperformance, another important element of the new approach is to recognize that more grant aid and a greater degree of concessionality in loan conditions is appropriate for countries that are prepared to reform but are still reliant on official financing. In part, this is a reaction to the adverse development status of many highly indebted poor countries. More generally, it reflects the fact that the resources required for rapid development exceed the level that can be comfortably borrowed and that excessive external debt distorts and constrains development strategies.

Another element of this initiative is to encourage renewed efforts at debt-reduction packages, both vis-à-vis the multilaterals but also with respect to bilateral creditors (e.g., through Paris Club agreements). This recognizes in part that previous financial assistance was perhaps inappropriately designed: a high debt burden reflects bad decisions by lenders as well as debtors. In addition, debt reduction is motivated by the to these countries to undertake a fresh start.19 This would represent a major change from current conventions that apply a set of uniform numerical criteria (e.g., with respect to GDP per capita, the debt-export ratio, and the ratio of private to official bilateral debt) in order to determine the scale of relief that is due each candidate country.20

As currently constituted, however, the IFIs cannot deliver all the assistance required by the CIS-7 in their development efforts. At a financial level, the resources controlled by the IFIs are limited, and deployment is subject to the constitutions and mandates of these organizations. As such, bilateral sources and global funds that aggregate donations from individual donors are important potential components of overall financial assistance. In this regard and in view of the competition for funds from other low-income countries, it is a challenge for the CIS-7 countries to demonstrate to such potential donors that they are indeed worthy recipients of assistance because each individual donor must weigh the marginal impact of extra aid in alternative destinations.

Of course, it is also desirable that donor nations increase the aggregate pool of funds potentially transferable to low-income countries such that all countries meeting reform criteria may receive access to sufficient funding. On this front, it is important that the donor countries strive to meet the aid targets set under the Millennium Development Goals. To ameliorate the proliferation problem, increased pooling of funds among donors is also to be encouraged. For the CIS-7 countries, further integration of aid policies by EU members would be particularly beneficial.

Catalyzing private financial flows is an important goal as well if sufficient is to be.21 In addition to helping governments create the institutional and infrastructural conditions for market-oriented growth, which is already the goal of the IFIs, this involves their becoming participant investors in collaboration with the private through various arrangements.22 Exploring an enhanced role for public-private partnerships and innovative ways by which the multilaterals can leverage private capital flows has also been supported by the United Nations.23

The complexity of the development challenges facing the CIS-7 countries is such that the scope of required policy advice and technical assistance is far broader than feasibly can be provided by the World Bank and, especially, the IMF. The consistency of approach required across member countries inevitably constrains the multilateral institutions in their ability to deviate too far from normal terms and conditions in individual countries. The legitimacy and universality of these institutions rely on taking an evenhanded approach. The financial structures of the institutions are also such that a higher volume of lending in one zone constrains the capacity to act in other regions that are deserving of assistance.24

Along these lines, the various UN agencies potentially have much to offer in specific reform areas, as do some nongovernmental organizations and EU programs. One implication is that, even if the two Bretton Woods institutions do a reasonable job in coordination, the problem of streamlining and managing the supply of policy advice and technical assistance from a broad set of agencies is, in itself, a considerable challenge. Moreover, for full effectiveness, an awareness of the activities and resource needs of all the other agencies must feed back into the operations and lending decisions of the IFIs.

Recent efforts to improve the targeting and better coordinate the delivery of technical assistance across the institutions are to be welcomed. This involves greater specialization in areas of core competence by each institution. Beyond the traditional function of training government officials, an interesting potential development would be to include representatives from civil society as participants in such training.25 An interesting historical parallel is that the Ford Foundation supported in the 1960s the development of a nongovernmental policy think tank in Ireland to establish an independent source of expertise on policy analysis. Such an initiative would have the effect of broadening the capacity for more effective program design and improving the quality of the PRSP process. Beyond the direct aid provided by the international community to the individual CIS-7 countries, their development has other international dimensions. Clearly, it is desirable that the European Union and other industrial countries further liberalize access to imports from these and other developing countries. Promoting cooperation among the CIS countries in reducing trade barriers and, where feasible, the cross-border integration of energy networks are positive sum endeavors.

Regional associations can also provide useful opportunities for information exchange, learning about reform experiments elsewhere, and enhancing cooperation on policy initiatives. Moreover, they can establish (formally or informally) feasible performance benchmarks that may stimulate reform efforts. Such yardstick competition generates peer pressure on laggards to improve performance.26 Although such countries do not form a contiguous geographical group, hopefully their participation in the CIS-7 initiative in itself can provide some of the benefits of membership in an international association. Enhanced international monitoring, cross-member demonstration effects, and the potential for information exchange on policy and institutional developments should all have beneficial effects on the quality of governance in these countries.

The prospects for development of the CIS-7 countries would also be greatly enhanced by a more rapid economic recovery in the collective CIS region, and Russia in particular. As such, an improved effort by the international community to assist the CIS-7 countries is best conducted in the context of a sustained strategy for the overall CIS. Of course, the needs of the different member countries vary greatly, and the nature of assistance may differ both in terms of intensity and form as between the low-income CIS-7 countries and other countries in the region. In turn, sustaining and deepening recent reforms in Russia would provide important leadership in demonstrating to the other CIS countries the viability and net gains to be achieved by embracing a more complete transition strategy.

Conclusions

Our goal in this paper has been to review and evaluate the performance of the international community and the IFIs in assisting the development of the CIS-7 countries, with an eye toward identifying the appropriate strategies for the future.

At least in some dimensions, the CIS-7 countries have made notable progress since the mid-1990s, and the IFIs have been heavily involved in efforts to encourage reform in these countries. The deeper reforms that remain, however, are primarily the responsibility of the domestic authorities: for instance, it would be an infringement on sovereignty for the IFIs to insist on particular designs for domestic public institutions and legal codes. If the international community wishes to intervene more directly in these areas, the IFIs are not the appropriate agents for such state building exercises.

In terms of policy advice, the initial focus was on securing macroeconomic stabilization, but the major priority has been to advance the reform agenda, especially in redefining the role of government in these countries. The resources committed to technical assistance have been substantial; however, the greatest success has inevitably been in the more technocratic policy areas, such as establishing functioning central banks, with relatively less penetration in the more diffuse areas, such as establishing the rule of law and creating an investment-friendly business environment. Related to the conclusion in the previous paragraph, this reflects the limits of the scope of the IFIs.

Financial flows from the IFIs to the CIS-7 countries have been very significant, as have bilateral flows in some cases. The large rise in official external debt relative to output during this period does not suggest an unwillingness to provide financing for these countries and may even represent overlending. In part, this signals that the institutions should become tougher in assessing whether financial support will be well used in each case. It also suggests, however, that greater concessionality is desirable for those countries that demonstrate a commitment to reform and development.

At this stage, many of these recommendations for change have already begun to be reflected in the official policies and practices of the IFIs and their major shareholders. In regard to the CIS-7 countries, perhaps the major challenge now is to ensure effective implementation of the new procedures and financial policies. Because all of these countries will likely remain clients of the IFIs in the medium term, this involves a sustained commitment to assist these countries in formulating and implementing appropriate development strategies. It also requires a determination to discriminate rigorously between progressive and antireform actions by the policymakers. In this way, by rewarding and encouraging progress in reform, the international community and the IFIs will continue to have an essential role in supporting the CIS-7 countries.

References

    ÅslundAndersPeterBoone andSimonJohnson (2001) “Escaping the Under-Reform Trap,”IMF Staff PapersVol. 48Special Issue pp. 88108.

    Braga deMacedoJorge (2002) “Comparative Development and Institutional Change,”mimeo,OECD Development Centre.

    BuiraAriel (2002) “An Analysis of IMF Conditionality,”Oxford University Department of Economics Discussion Paper No. 104.

    BuiterWillem andHans PeterLankes (2000) “IFIs As Partners of the Private Sector,”mimeo,EBRD.

    CasellaAlessandra andBarryEichengreen (1996) “Can Foreign Aid Accelerate Stabilization?”Economic JournalVol. 106 (May) pp. 60519.

    DeMeloMartha and others (2001) “Circumstances and Choice: The Role of Initial Conditions and Policies in Transition Economies,”World Bank Economic ReviewVol. 15 pp. 131.

    EasterlyWilliam (2001) The Elusive Quest for Growth: Economists’ Adventures and Misadventures in the Tropics (Cambridge, Massachusetts: MIT Press).

    FalcettiElisabettaMartinRaiser andPeterSanfey (2002) “Defying the Odds: Initial Conditions, Reforms, and Growth in the First Decade of Transition,”Journal of Comparative EconomicsVol. 30 (June) pp. 22950.

    FischerStanley (2001) “Ten Years of Transition: Looking Back and Looking Forward,”IMF Staff PapersVol. 48Special Issue pp. 18.

    GuptaSanjeev and others (2002).Is the PRGF Living Up to Expectations? An Assessment of Program Design IMF Occasional Paper No. 216 (Washington: International Monetary Fund).

    IMF and World Bank (2002) Strengthening IMF-World Bank Collaboration on Country Programs and Conditionality – Progress ReportWashington, DC.

    KapurDevesh (2002) “Do As I Say Not As I Do: A Critique of G-7 Proposals on Reforming the Multilateral Development Banks,”Center for Global Development Working Paper No. 16.

    MukandSharun andDaniRodrik (2002) “In Search of the Holy Grail: Policy Convergence, Experimentation, and Economic Performance,”NBER Working Paper No. 9134 (WashingtonNational Bureau of Economic Research).

    RolandGerard (2001) “Ten Years After—Transition and Economics,”IMF Staff PapersVol. 48Special Issue (Washington: International Monetary Fund) pp. 2952.

    SachsJeffrey (1996) “Reforms in Eastern Europe and the Former Soviet Union in Light of the East Asian Experiences,”NBER Working Paper No. 5404 (Washington: National Bureau of Economic Research).

    SachsJeffrey (2002) “Resolving the Debt Crisis of Low-Income Countries,”Brookings Papers on Economic ActivityVol. 1 (Washington: Brookings Institution), pp. 25786.

    TornellAaron andPhilip R.Lane (1999) “The Voracity Effect,”American Economic ReviewVol. 89 (March) pp. 2246.

    United Nations (2002) Report of the International Conference on Financing for DevelopmentNew York.

    WolfHolger (2002) “Policy Choices and Outcomes: Initial Conditions in the Economics of Transition,”mimeo,Georgetown University.

    World Bank (1998) “Assessing Aid: What Works, What Doesn’t, and Why,”World Bank Policy Research Report (Washington: Oxford University Press).

    ZecchiniSalvatore (1995) “The Role of International Financial Institutions in the Transition Process,”Journal of Comparative EconomicsVol. 20 (February) pp. 11638.

The author is affiliated with the Institute for International Integration Studies (IIIS) and Economics Department, Trinity College Dublin, and Centre for Economic Policy Research. Colin Andrews and Charles Larkin provided helpful research assistance, and David Robinson at the IMF provided assistance in sourcing materials and arranging interviews. The author is grateful to those IMF staff members who shared their views with him and also to Holger Wolf and other participants in the preconference workshop in Washington on October 17, 2002. This work was supported by the IIIS.

The text is available at the G-7 website maintained by the University of Toronto: http ://www. g7. utoronto .ca.

See Roland (2001) for a presentation of the gradualist viewpoint.

Against this view, Fischer (2001) points out that this approach assumes that there exists a degree of political stability and a capacity to forecast the impact of specific reforms, which was absent in many transition countries, most particularly in the CIS.

For instance, Sachs (1996) argues that a big bang approach was desirable for countries in Eastern Europe and the former Soviet Union that were highly urbanized, with extensive social welfare systems and large state-owned industrial sectors that needed rapid dismantling.

The explanatory power of these regressions is typically low. This is not surprising because we know that set of transition countries varies enormously across a range of dimensions. See de Melo and others (2001) and Falcetti, Raiser, and Sanfey (2002) for recent analyses of the determinants of reform efforts. Our purpose is merely to document a group difference in average reform efforts.

The EU through its TACIS program, some United Nations (UN) agencies, and bilateral donors have also provided technical assistance in specific policy areas.

In some cases (e.g., Tajikistan in 1993), the World Bank provided grant aid to assist the government in managing and making effective use of external assistance. See also Zecchini (1995).

See http://www.clubdeparis.org for details. Debt-equity swaps and other conversion arrangements have also been part of these debt renegotiations.

This is labelled the “voracity” effect by Tornell and Lane (1999): their theoretical model predicts that an increase in aid also generates an increase in other borrowings.

As per Wolf (2002).

See also Buira (2002) on the evolution of IMF thinking regarding the conditionality debate.

See the IMF and World Bank (2002) for an analysis of the scope for improved coordination between the two institutions.

Indeed, the Meltzer Commission recommended the abolition of the PRGF at the IMF.

Easterly (2001) provides a sharp critique of the lending culture of the World Bank and the poor returns on many of the supported projects.

Sachs (2002) argues that the terms of debt relief should be considered on a country-by-country basis, using in-depth analysis of each debtor’s specific profile and prospects.

The most recent Paris Club agreement for a CIS-7 country was For the Kyrgyz Republic in March 2002, which qualified for “Houston terms,” providing an intermediate level of favorable treatment.

Buiter and Lankes (2000) argue that the constrained resources of the IFIs means that their mandates should be reoriented toward facilitating private flows to low-income countries.

Zecchini (1995) highlights the potential of investment partnerships with the private sector.

Kapur (2002) emphasizes that the IFIs do not have the competence nor the flexibility to provide assistance on all aspects of the institution-building process that must take place in such countries.

See also Braga de Macedo (2002) on the potential of regional arrangements in stimulating reform and development efforts.

    Other Resources Citing This Publication