Chapter

Chapter 2 The Design of Jordan’s IMF-Supported Programs

Author(s):
International Monetary Fund. Independent Evaluation Office
Published Date:
April 2006
Share
  • ShareShare
Show Summary Details

The main question discussed here is how well suited to addressing Jordan’s major adjustment challenges were the objectives, strategies, and policies contained in the various programs supported by the IMF. The chapter also briefly reviews the nature of the IMF’s analysis and policy advice following Jordan’s “exit” from IMF arrangements, in the context of post-program monitoring discussions.

Specifically, the focus is on the clarity and coherence of the strategies and policies that were expected to contribute to external sustainability, growth, and inflation:

  • How restoring external and fiscal sustainability was analyzed and defined and how this influenced program design.

  • How the pace and composition of targeted macroeconomic adjustment were derived and whether they were consistent with the broader objectives of the programs. In this context, particular attention is paid to the scale and composition of the targeted fiscal adjustment and the policies expected to achieve this adjustment.

  • Inflation objectives and the design of monetary and exchange rate policy.

  • The structural reform content of the various programs and whether this content was well suited to helping Jordan resolve its key adjustment challenges.

  • The scope and content of program conditionality.

Several points are worth noting at the outset. First, it is not intended to report here a detailed blow-by-blow analysis of the contents of each individual program, especially since details of program design were modified during program reviews as circumstances evolved.

Second, the review covers a 15-year period during which the IMF’s approach to some aspects of program design was modified as a result of various IMF-wide reviews that sought to derive lessons for enhancing the overall effectiveness of programs. The most relevant of these for the case of Jordan were the 1994 and 2000–02 “conditionality” reviews. A key message from the former was the need for programs to set fiscal adjustment explicitly in a medium-term framework that allows for evaluation of progress toward sustainability. The 2000–02 review dealt mainly with structural conditionality; it called for the “streamlining” of such conditionality in terms of quantity, scope, and criticality for meeting program objectives.1 In this sense, there have been some changes in the internal measuring rods by which appropriate program design was judged and it is important not to judge earlier programs against explicit guidelines that were only established later.

Finally, any assessment of the appropriateness of program design has to take account of the likelihood that it will be implemented. A program that looks strong and internally consistent on paper but that has little domestic ownership is unlikely to be successful. This issue is pursued further in the next chapter.

Approaches to External and Fiscal Sustainability in Program Design

A review of program documents suggests several conclusions about the approach to restoring external and fiscal sustainability incorporated into program design in Jordan.

The operational approach to external viability that effectively drove program design was a flow, not a stock, concept—at least until very recent programs. Restoration of medium-term external viability was defined in terms of eliminating the need for “exceptional financing” (i.e., nonautonomous flows including debt relief, arrears accumulation, or IMF financing). While medium-term projections of the balance of payments prepared at the time of the programs included debt projections, there was, until the last few years, much less focus on the analysis of debt dynamics as a driver of the magnitude of targeted adjustment in the external current account. This analysis was largely limited to (1) a qualitative judgment that Jordan suffered from a “debt overhang” so that lowering external debt ratios would be good for growth by reducing potential disincentives for the private sector to save and invest; and (2) a judgment that a declining path of debt/GDP and debt/foreign exchange earnings indicated “sustainability,” without any assessment of what would be appropriate longer-term targets and why or how such judgments might influence the pace of targeted adjustment in, say, the external current account.2

Projections of when such exceptional financing would be eliminated were ambitious and proved to be consistently optimistic (Table 2.1). The reasons for this are hard to disentangle, but a reluctance to acknowledge initially how long the adjustment process would be (and, correspondingly, the time span over which debt relief would be required) played a role. However, after a series of flow reschedulings from the Paris Club, the combination of an improving balance of payments situation as well as an innovative approach to an “exit” rescheduling (and the related signaling from the IMF) allowed Jordan to phase out such financing (Box 2.1).

Table 2.1.Target Dates for Elimination of Exceptional Financing Needs1
Date Specified in Original

Program Document
1989 SBA1995
1992 SBA1998
1994 EFF1998
1996 EFFBeyond 20022
1999 EFF2002
2002 SBA2007
Source: IMF program documents.

The approach to fiscal sustainability that underlay program design followed a similar pattern. The focus in early programs was on tailoring the targeted fiscal deficits to available financing. The early SBAs defined fiscal sustainability in terms of reducing initial large deficits to levels that could be financed by “normal” aid flows and domestic financing that was consistent with internal financial stability.

A shift to a broader concept of sustainability, including a more comprehensive analysis of the tradeoffs involved in various debt paths, would have been useful much earlier than it actually occurred. An approach to sustainability that concentrated on the viability of the financing of external and fiscal deficits in the short term was understandable in the early programs where a key concern was to ensure that projected financing gaps could actually be covered. However, by the time of the EFF arrangements, program design would have benefited from greater inputs from an analysis of debt dynamics to illustrate, for example, the trade-offs involved in choices on the magnitudes of targeted adjustment.3 As will be discussed further below, the focus on shorter-term sustainability issues also limited the analysis of potential growth consequences of program design.

Nevertheless, a marked shift in the depth of the debt sustainability analysis (DSA) underlying program design in Jordan did take place beginning in 2002 with the preparation of a high-quality medium-term fiscal strategy paper. The shift also reflected the adoption, IMF-wide, of a more comprehensive approach to DSA that called for systematic medium-term debt projections and assessments of the impact of the most relevant shocks on the debt profile.4 The shift to a greater focus on targets for the debt stock as a key input into program design tracked Jordan’s adoption of explicit ceilings on total public and external debt as part of a new Public Debt law (Box 2.2).

Did these recent sustainability analyses contain an adequate assessment of the potential vulnerabilities facing Jordan’s economy, and how did they feed into program design and the IMF’s macroeconomic policy advice in the post-program period? Three key findings emerged from a review of the various sustainability analyses undertaken since 2002 (see Annex 3 for further details).

  • The sustainability analyses did pay increasing attention to what would be an appropriate longer-term debt target for Jordan. Moreover, while there had been relatively little analysis in early program documents of the consequences for growth of different paths of external adjustment, IMF staff did make greater efforts to assess such potential trade-offs from 2002 onward.

  • The analysis of potential trade-offs between different strategies for continuing progress toward longer-term sustainability did result in a significant shift in program design—especially between the 1999 EFF and the 2002 SBA. The 2002 fiscal strategy paper concluded that the fiscal strategy underlying the EFF needed substantial modification in favor of a revised approach to addressing the debt problem that emphasized (1) a more gradual fiscal deficit reduction with more emphasis on specific longer-term reforms to address key fiscal rigidities; and (2) more aggressive “below the line” efforts to reduce the debt stock through debt restructuring/reduction and proceeds from privatization.

  • The treatment of grants became a critical factor in judgments of longer-term sustainability. Assessments in Board papers rightly emphasized that the temporary upsurge in grants in 2003–04 should not be used to finance a permanent pickup in government expenditures. However, the assessments said relatively little about the potential dangers involved in managing a sharp reversal of grants or about policies that could help smooth the macroeconomic consequences. Internal staff notes tended to be somewhat franker about the risks, and information gathered from interviews for this evaluation suggest that the potential risks were discussed with the authorities.

Box 2.1.Paris Club Rescheduling and IMF Signaling

Jordan has had six rescheduling agreements with the Paris Club facilitated by the country’s six IMF arrangements. Typically, the agreements provided relief on qualifying debt service falling due during periods that roughly corresponded to that of existing IMF arrangements.

All together, nearly $5 billion of debt-service obligations were rescheduled under the six agreements. At the time of the 2002 agreement, about 60 percent of the country’s outstanding public and publicly guaranteed external debt was due to Paris Club creditors.

All the agreements have been on nonconcessional terms, on account of Jordan’s status as a middle-income country. However, over time, the terms became more favorable, including longer repayment and grace periods for the rescheduled amounts. For example, under the 1997 and subsequent agreements, payment on official development assistance (ODA) loans were rescheduled over 20 years with a 10-year grace period; payments for non-ODA debt were rescheduled over 18 years with a 3-year grace period.

The 2002 agreement, which was designated an “exit” agreement, had the unusual feature of having a consolidation period that extended well beyond the expected expiration of the 2002 SBA, covering debt-service obligations falling due through 2007. As Jordan was expected to graduate from reliance on IMF arrangements at the end of the 2002 SBA, the agreement indicated that after such a graduation, debt relief would be provided to Jordan on the basis of “strong IMF involvement through semiannual monitoring of Jordan’s performance.” In the event, IMF “involvement” has been through the post-program monitoring (PPM) mechanism which is applied to countries that have graduated from IMF arrangements but which have obligations to the IMF in excess of 100 percent of their quota. The Executive Board completed the first PPM discussions on Jordan in January 2005.

Box 2.2.Public Debt Law of 2001

In 2001, Jordan enacted a new Public Debt Law with the twofold goal of providing a more effective framework for debt management and facilitating a medium-term strategy for the reduction of dependence on external financing. The law required that total public debt be limited to 80 percent of GDP and that external debt be limited to 60 percent of GDP by 2006. It also removed limits on the government’s ability to finance fiscal deficits through issuance of domestic debt rather than through external borrowing. The law limits internal borrowing to debt instruments (bills and bonds) traded on the Amman stock exchange and prohibits direct borrowing from commercial banks or other institutions.

In the context of the 2002 SBA, the authorities presented a medium-term fiscal strategy for debt reduction designed to achieve the target set out in the law.

Growth Objectives and the Magnitude and Composition of Targeted Adjustment in Key Macroeconomic Variables

The broad strategy to foster sustainable growth over the medium term was in many respects unchanged throughout the long sweep of IMF program involvement, reflecting the authorities’ underlying policy framework.5 Specific growth projections varied substantially—partly reflecting the consequences of various geopolitical shocks—and macro-economic targets as a share of GDP exhibit discontinuity because of significant revisions to national accounts data.6 However, the core elements of the medium-term strategy were generally as follows under all the arrangements:

  • The private sector was expected to be the primary engine of growth, supported by a policy framework that provided appropriate incentives for increased private investment and savings.

  • A significant decline in public sector dissaving and the public debt “overhang” (along with the associated restoration of external viability and lower inflation) would help provide those necessary incentives.

  • The increases in overall domestic investment would necessarily be fairly moderate and the contribution to be financed by domestic savings would rise markedly. This reflected the sharp targeted adjustment in the external current account (excluding grants) that continued at least through the 1999 EFF (Figure 2.1 and Annex 13, Table A13.2).

  • Given the moderate increases in investment, a significant part of any pickup in growth over the medium term would have to come from higher productivity growth. Structural reforms designed to achieve this included liberalizing domestic trade and domestic prices; a wide range of institutional and policy reforms aimed at enhancing the business and investment climate; and a reduction of the state’s intervention in the economy, including through privatization of state-owned enterprises.

Figure 2.1.Medium-Term Targets for External Current Account and Projected Evolution of External Grants 1

(In percent of GDP)

Source: IMF program documents.

1 The numbered arrows show the projected medium-term path at the time of the original program design, beginning with the then-estimated actual level in the year before the program was formulated (t–1). The solid line shows latest estimated actual outturns. Some revisions to the (t–1) estimated actuals occurred after the programs were formulated, in which case the arrows showing the direction of targeted adjustment do not start on the actuals line.

2 Balance of payments definition.

While the qualitative nature of the strategy (and the expected evolution of savings and investment balances) was spelled out quite clearly in program documents, there was little analysis of what this strategy might mean in more concrete terms for the medium-term growth projections. For the most part, these projections appear to have been arrived at exogenously (e.g., relying on extrapolations from recent trends and objectives established in the authorities’ own strategy documents). The general pattern was to project a rapid recovery of growth when output performance was poor at the start of the program and to project the continuation of a good growth performance if that was already the situation. Program documents would have benefited from more thorough analyses of the sources of growth and of the impact of envisaged fiscal adjustment on private savings and investment. Clearly, all such analytical exercises have their limitations, but spelling out the assumptions underlying some of these key elements of program design would have facilitated judgments on their reasonableness and on how programs might need to be modified should the assumptions prove incorrect. As noted above, there has been significantly more analysis of these issues in recent years.7

Within this broad strategy, however, there were some significant differences in approach between arrangements with respect to how the targeted external current account adjustment was to be achieved (Table 2.2 and Figure 2.2). In summary, the early SBAs expected all of the adjustment to be accounted for by a sharp reduction in the fiscal deficit (excluding grants). For reasons that are not well explained in program documents, the 1994 EFF, which targeted continued significant adjustment in the external current account, expected most of it to be generated by a higher private savings-investment balance. By the time of the 1999 EFF, when the adjustment in the external current account had essentially been achieved, the programs targeted a large—and, in retrospect, unrealistic—further reduction in the fiscal deficit in order to create room for a substantial “crowding in” by the private sector. In the event, none of the fiscal adjustment targeted under the three EFFs was achieved. The strategy underlying the 1999 EFF was eventually replaced by one under the 2002 SBA involving a more moderate path for the deficit, with the aim of greater focus on underlying fiscal reforms. However, the upsurge in grants that began in 2003 eventually led to a further change in strategy, with a significant proportion of the increased resources expected to be absorbed by higher fiscal deficits (excluding grants) but with part expected to go to financing higher private investment.

Table 2.2.Some Stylized Facts About the Composition of Targeted Adjustment
ArrangementSize of Targeted External Adjustment and How It Was to Be Achieved1What Actually Happened2
1989 SBA
  • Large reduction in external current account deficit (by 9.2 percent of GDP).

  • All from higher domestic savings; no change in investment.

  • All from fiscal adjustment, mainly from expenditures.

  • Current account deficit originally widened in 1990–91, under impact of the war in the Middle East, but then steadily declined.

  • Fiscal deficits also widened initially but then massive fiscal consolidation took place in 1992—exceeding the cumulative targets under the two SBAs.

  • Over period 1988–92, all of adjustment came from fiscal side.

  • The actual decline in the private sector savings-investment balance while significant was much less than programmed.

1992 SBA
  • Large reduction in external current account deficit (12.7 percent of GDP).

  • All from higher domestic savings; investment higher.

  • Two-thirds from fiscal adjustment, mainly from expenditures.

1994 EFF
  • Large reduction in external current account deficit (8.7 percent of GDP).

  • Three-fourths from higher savings; one-fourth from lower investment.

  • One-third from fiscal adjustment, all from expenditures.

  • Current account deficit continued to narrow steadily through 1999 despite the fact that none of the targeted fiscal deficit reduction under the three EFFs was achieved.

  • Public debt ratios continued to decline despite fiscal slippages because of debt reduction operations.

  • In contrast to program objectives, virtually all of the adjustment in the external current account during the period of the three EFFs reflected an improvement in the private savings-investment balance.

1996 EFF
  • Modest reduction in external current account deficit (2.6 percent of GDP).

  • All from higher savings; investment slightly higher.

  • Virtually all from fiscal adjustment; all from expenditures.

1999 EFF
  • No significant change in external current account.

  • Significant increase in domestic savings to finance higher investment.

  • Large fiscal deficit reduction (i.e., large implicit “crowding in” of private sector).

  • Two-thirds of fiscal reduction from expenditures

2002 SBA
  • Moderate widening of external current account deficit (by 1.7 percent of GDP).

  • Savings broadly unchanged; moderate increase in investment.

  • Moderate widening of fiscal deficit, financed by higher grants.

  • Most of targeted widening of deficit reflected revenue reduction.

  • External current account and fiscal deficits widened more than targeted, reflecting a sharp upsurge in grants that began in 2003.

2004 post-program monitoring
  • Large (temporary) widening in the external current account, financed by sharply higher grants.

  • Lower savings and higher investment.

  • Both fiscal deficit and private sector savings-investment balance projected to deteriorate.

Figure 2.2.The Envisaged Adjustment in Fiscal Balance, Excluding Grants

(In percent of GDP)

Sources: IMF program documents and IMF (2004b).

Within the targeted fiscal adjustment, expenditures were typically expected to bear most or all of the burden (see Table 2.3), but program documents did not focus enough on the fiscal structural reforms that would be needed to sustain this adjustment over the medium term. The focus on expenditure reduction was appropriate, given the initial high expenditure to GDP ratio and the need to alter the composition of revenues for efficiency reasons. However, programs would have benefited from a greater emphasis on identifying the specific reform steps to address the underlying rigidities. The following broad conclusions are suggested by a review of the fiscal structural content of various arrangements (see Annex 5 for details):

  • The emphasis of the first SBA was to help the fiscal situation by adjusting administrative prices, reducing untargeted food subsidies, and increasing specific excise taxes. Given the crisis atmosphere and the need for a rapid fiscal adjustment, it is unrealistic to expect that a comprehensive program of structural reform could have been put together at this time.

  • The EFFs started identifying some structural measures and supported steps in the areas of pensions, better targeting of food programs, and divestiture and restructuring of public enterprises. However, they did not address other major rigidities on the expenditure side. In particular, little was done to address, even in a long-term framework, the issues of the large public sector wage bill. The EFFs spanned a seven-year period, a sufficient period under which some initial steps could have been identified in this area—given the emphasis that program documentation gave to the expenditure rigidity issue.8

  • As will be discussed further in Chapter 5, cooperation between the IMF and World Bank on public expenditure reviews was not very effective in generating concrete plans of action for expenditure restructuring during this period.

Table 2.3.Medium-Term Fiscal Objectives and Projections of Different IMF-Supported Programs and Papers(In percent of GDP)
YearFiscal Balance

After Grants
GrantsFiscal Balance

Before Grants
RevenueExpenditure
Initial values (1988, as seen at the time)–16.47.323.731.855.5
The early SBAs
1989 SBA1991–8.06.0–14.034.048.0
1993–6.03.0–9.036.545.5
1992 SBA1995–6.71.8–8.528.036.5
1998–3.61.4–5.030.035.5
The EFFs
1994 EFF1995–0.93.4–4.331.936.2
19980.22.7–2.530.833.3
1996 EFF19980.12.6–2.530.332.8
2000–1.00.8–1.830.332.1
1999 EFF2001–1.42.6–4.030.534.5
The 2002 SBA
2002 SBA2003–4.36.0–10.325.235.5
2007–2.73.6–6.324.731.0
2004 monitoring paper12005–3.29.9–13.125.238.3
2007–2.51.2–3.724.728.4
Source: IMF program documents.

The 2002 fiscal strategy paper did lay out a comprehensive policy agenda to support the adjustment path it proposed. Progress was made in further supporting pension reforms and tax reforms in the sub-sequent 2002 SBA, but a number of other key expenditure rigidities remain.

Moreover, the potential challenges for fiscal adjustment posed by the expected reversal of the recent large upsurge in grants have received insufficient attention in recent Board papers. As a result of developments in the Middle East, including the Iraq war, Jordan began to receive a sharply increased level of grants that were reflected in a more than doubling of the targeted fiscal deficit before grants to about 10 percent of GDP under the 2002 SBA.9 Since the surge in grants was expected to be sharply reversed after a few years, the 2002 SBA documentation incorporated projections for the medium term that rapidly reversed the fiscal expansion programmed for 2003, to be achieved through a projected contraction in public expenditures from 35.5 to 31 percent of GDP by 2007 (Table 2.3). The projected swings in the fiscal stance became even sharper in the macro-economic framework presented in the 2004 post-program monitoring paper that discusses Jordan’s remaining adjustment challenges as it exits from IMF financial assistance. That paper envisages a peak level of grants for 1995—reaching almost 10 percent of GDP—and a peak level of public expenditures equal to 38 percent of GDP—the largest level during the last decade. The same paper projects then an extraordinary level of adjustment between 2005 and 2007: grants are to decline to a very low level of 1.2 percent of GDP, and public expenditures are to contract to a historically low level equal to 28 percent of GDP (Figure 2.3). In spite of this extraordinary fiscal adjustment being projected—10 percentage points of GDP in a two-year period—the post-program monitoring paper does not alert the authorities about this impending fiscal correction ahead of them and the urgent need to identify policy options to confront it.

Figure 2.3.Envisaged Adjustment in Public Expenditures 1

(In percent of GDP)

Sources: IMF program documents and IMF (2004b).

1 Each observation represents a program target. There are multiple observations for some years because of overlapping projection periods under some of the arrangements.

Inflation Objectives and the Design of Monetary and Exchange Rate Policy10

The inflation objectives of the various programs were straightforward. The first two SBAs aimed to bring down inflation quickly (from about 26 percent in 1989 to around 6–7 percent by 1993–94). They succeeded, and subsequent arrangements aimed to keep inflation low (typically in the range 2–4½ percent a year; see Annex 13, Table A13.2).

The authorities’ framework for conducting monetary policy changed markedly, reflecting shifts in the choice of exchange rate regime in the early years of IMF program engagement (see Annex 6 for details), and subsequently, a move from direct credit controls to indirect monetary control. The key elements of program design were as follows:

  • Following the effective floating and significant depreciation of the Jordanian dinar in late 1988, exchange rate policy was characterized as one of “managed flexibility” during the early 1990s. What this actually meant in terms of a nominal anchor for monetary policy was not very transparent, especially since some statements about exchange rate policy in program documents seemed to suggest a real effective target.11 In practice, the rate was held stable against the SDR basket of currencies for much of this period, which proved effective in moderating inflation. At the same time, policy instruments still relied heavily on various direct controls, including of various exchange transactions and tight direct credit control measures.

  • Exchange rate policy shifted to a fixed peg to the U.S. dollar—on a de facto basis in about 1994 (in the sense that the exchange rate has been essentially stable vis-à-vis the U.S. dollar since then) and in a formally announced manner from October 1995 onward. Thereafter, the principal objective of monetary policy became defending the peg, mainly through a policy of maintaining the attractiveness of dinar-denominated certificates of deposits vis-à-vis U.S. treasury bills. The authorities acted on a number of occasions—with the encouragement of IMF staff—to widen this interest rate spread in order to prevent a shift out of dinars during periods of uncertainty (e.g., during the period of the illness and death of H.M. King Hussein).

  • The framework for designing the monetary program under IMF arrangements remained virtually unchanged; it revolved around a ceiling on net domestic assets.12 This was the traditional IMF approach and inevitably involved something of a disconnect between the principal policy instrument (namely, interest rates) and the formal structure of the monetary program (namely, quarterly net domestic assets (NDA) targets). This disconnect is not necessarily a problem and the framework could be interpreted as a “belt and braces” approach whereby the quantitative targets act as a potentially useful signal of emerging problems especially since it is probably not practical to have formal conditionality on day-to-day interest rate policy.13 Much depends on its implementation in practice, since estimates of underlying behavioral parameters—especially the demand for money—are inevitably uncertain. If applied in an overly rigid manner, it could lead to an “overdetermined” system.14 Any deviations from NDA targets would need to be judged in the context of whether interest rate policy is being conducted appropriately and the objectives for inflation and external reserves are being achieved. This flexible approach does appear to have been applied in the case of Jordan. Moreover, recent Board papers placed the primary emphasis on discussion of interest rate policy.

Structural Reforms Addressed by IMF-Supported Programs

The IMF and the World Bank collaborated closely on structural reforms in Jordan, with the World Bank taking the lead in several areas including the trade regime and regulatory and legislative reform for enhancing the investment climate and promoting private sector development. There were also many common areas of work, such as trade, energy, price deregulation, and privatization—areas where the Bank was also active with its own lending operations. The areas that were emphasized in the IMF-supported programs are covered below.

Tax reform. The main objectives were making revenues more buoyant and reducing the distortionary nature of the tax system. The centerpiece of the reforms was the reorientation of the tax system from an import-based one to one based on domestic consumption. In particular, a general sales tax (GST) with the features of a value-added tax (VAT) was to replace a series of existing narrowly based consumption tax. Other reforms were geared to broaden the base of direct taxes and to strengthen revenue administration. Tax reforms were closely coordinated with tariff reforms—a key element of reform of the trade regime.

Trade regime. Reform in this area figured predominantly under the first SBA and in a 1990 World Bank adjustment loan. The reforms followed the well-established principle of first converting quantitative restrictions into tariffs and replacing specific tariffs by ad valorem ones. Thereafter, effective protection was gradually lowered through the 1990s supported by the EFFs. Maximum tariffs were reduced from above 300 percent to 50 percent between 1989 and 1996, and then to 35 percent in 1999.

Price liberalization and targeting of food subsidies. The adjustment of controlled prices during the early 1990s was accompanied by the relaxation of restrictions on the import and export of food commodities by the private sector, thus increasing the role of market mechanisms in determining prices. Such reforms were part of a broader reform to replace general food subsidies by more targeted subsidies—as the former had favored the better-off more than the poor.15 Eventually, food subsidies were gradually replaced by direct cash assistance administered by the National Aid Fund (NAF).

Financial sector. Financial sector reforms included strengthening the banking system, improving the national payment system, and deepening financial markets. IMF involvement was largely through an intensive program of technical assistance. Specific areas of support included enhancing the Central Bank of Jordan supervision capacity and bringing reporting practices to international standards. Transformation of monetary operations from direct to market-based indirect instruments was also a key element of financial sector reforms. The World Bank complemented these activities through its Economic Reform and Development Loans of 1997 and 1999, by incorporating many actions on decontrol of interest rates and new banking legislation.

Budget management. Reforms in this area started late, gaining momentum after the problems with data revisions in 1998 that led to the interruption of the 1996 EFF. Among the most significant reforms was the establishment of a macrofiscal unit in the Ministry of Finance to compile and analyze monthly fiscal data, and the introduction of a single treasury account system to improve the accuracy and timeliness of reconciliation and reporting of the state of public finances. More recently, the IMF and the World Bank have been collaborating on public expenditure management reforms aimed at improving budget formulation and execution, including introducing a medium-term framework.

Pension reform. Reforms in the public pension system—specifically, the pension schemes for the civil service and the military—were needed to address rapidly growing pressures on the fiscal position but were only addressed significantly beginning with the 1999 EFF and with technical assistance from both the World Bank and the IMF. The strategy had two major components: First, the civil service pension system was supplanted by the private pension system administered by the Social Security Corporation (SSC), with the objective of gradually phasing out the civil service pension scheme. Second, the authorities closed the military pension system to new entrants, completing the integration of the public and private sector pension systems.

Privatization and improvements in business environment. Privatization became an important part of the authorities’ strategy from 1996. Its purpose was to enhance efficiency, attract investors, promote capital market development, and help public finances. It was complemented with legislation to strengthen the judiciary and regulatory agencies, regulate leasing, and abolish controls on limits to foreign ownership. This was an area supported heavily by the World Bank through its adjustment operations. IMF involvement was more limited, focusing on the divestiture of some specific enterprises and an improved regulatory framework for the power sector to facilitate the entry of the private sector.

Design of Conditionality

Macroeconomic conditionality

Macroeconomic conditionality in all the programs sought to place (quarterly) limits on (1) some measure of domestic credit expansion; (2) the government budget deficit or some component of its financing; and (3) contracting of new nonconcessional public debt. Formal conditionality on targets for international reserves was introduced with effect from the 1994 EFF (Annex 7). Potential tensions between the structure of conditionality with regard to the monetary program and the changing nature of Jordan’s monetary policy framework have already been noted. Here, the focus is on a second issue. For a country in Jordan’s situation, how should grants be treated in fiscal conditionality?

This is a complex issue and it is useful to distinguish between considerations for determining the appropriate medium-term fiscal path and those involved in setting quarterly performance criteria. Clearly, the sustainability and predictability of the level of grants is a key consideration for both aspects. The IMF’s own approach on the issue is that, if a country can reasonably be expected to rely upon a particular level of grants over the medium term, then the sustainable path for fiscal policy can be derived using the deficit including a realistic projection of grants. It was this consideration that led to the switch in the performance criterion on the fiscal deficit from a “before-grants” to an “aftergrants” measure in the 2002 SBA. But the subsequent surge in grants in 2003–04 suggests two additional messages:

  • In Jordan’s case, it was prudent and realistic to formulate a medium-term fiscal path by analyzing both concepts and their longer-term implications. The “after-grants” measure indicated the impact on the public debt dynamics, whereas the “before-grants” measure flagged the magnitude of adjustment that might be required if grants could not be sustained. In practice, the staff did analyze both concepts in making their judgments about the appropriate fiscal path but, as already noted, this did not translate into sufficiently strong messages about the challenges of handling expected future sharp declines in grants.

  • While quantitative conditionality has a significant role to play in monitoring and signaling, the importance of such formal conditionality should not be exaggerated, since it is no substitute for a broader assessment of the sustainability of policies. Any single deficit measure has considerable limitations as a short-term performance criterion (PC) when the level of grants is unstable and needs to be interpreted with caution. An “after-grants” PC can lead to overly optimistic signals that the program is “on track” if there is an unexpected temporary pickup in grants—which may have been a factor in the case of the 2002 SBA.16 So the key message is not to be overly reliant on such conditionality alone. The main focus has to be on the implications for the medium-term fiscal path and debt dynamics.

Structural conditionality

Two aspects of structural conditionality are reviewed in this section. First, trends in the number of structural conditions are discussed while recognizing that such “head counts” provide only limited information on the overall burden and feasibility of conditionality. The discussion then turns to the coverage and content of such conditionality.

Compared with IMF-wide averages for SBAs and EFFs in nontransition countries, the initial program years under the 1989 and 1992 SBAs each had an above-average number of formal structural conditions (Annex 13, Table A13.6); subsequent years under the programs had about the average number of conditions. The conditions were mainly prior actions on increases in a number of administered prices and excise taxes. By contrast, the 1994 and 1996 EFFs had little formal conditionality. During 1994–98, the average number of structural conditions each year in Jordan’s programs was under three, compared with an average of over seven IMF-wide. In contrast, there was a marked escalation of conditionality under the 1999 EFF. Under the 2002 SBA, formal conditionality was again reduced, reflecting the IMF-wide initiative to streamline structural conditionality.

In terms of coverage and content, structural conditionality tended to match the structural policy content of programs. Initially, it focused primarily on specific pricing actions and then, beginning with the 1992 SBA and stretching through all of the subsequent arrangements, various aspects of tax reform (see Annex 10). In contrast, conditionality directed at addressing structural rigidities on the expenditure side was quite limited until the 1999 EFF—with the main exception being actions to contain food subsidies. Pension reform and improved budget monitoring and control procedures were the subject of conditionality in the 1999 EFF and the 2002 SBA, but steps to address other key fiscal rigidities—such as the overall wage bill—were never the subject of conditionality, reflecting the political sensitivities discussed earlier. From 1999 onward, there was also a greater focus on conditionality with regard to the financial sector.

Most of the structural conditionality—with the exception of the 1999 EFF—reflected a strong emphasis on respecting a clear division of labor with the World Bank—well in advance of the broader initiative to “streamline” such conditionality. For the most part, there was no attempt to duplicate conditionality in areas of focus of the World Bank, although program documents reflected the broader agenda being supported by the two institutions. The only exception was the inclusion of conditionality on tariff reforms as prior actions starting with the 1994 EFF, and the award of a management contract for Amman Water as a performance criterion in the 1996 EFF, both in areas that were being handled by the World Bank. Since these two actions were a part of the Bank’s program, the rationale for including them in the IMF-supported program was not strong.

The marked exception to this relatively “streamlined” approach was the first two annual programs under the 1999 EFF, which contained three to five times as many structural conditions as any preceding program—with a heavy emphasis on the restructuring, including privatization, of state enterprises. The reasons for this are not explained in the program documents but discussions with IMF staff suggest that one motivating factor was a desire to accelerate structural reforms in some areas where progress had been judged to be slow. But, many of the detailed conditions were not well designed. In particular, 14 conditions (mainly structural benchmarks), out of a total of 36 under the EFF, were related to different aspects of privatization. These 14 conditions included a rather diffuse package that appeared to pick out a number of actions from the ongoing privatization program supported by the World Bank (some of which, according to Bank staff, had already been implicitly discarded by the World Bank as not likely to be effective). They suffered from several design problems: (1) several conditions were in the form of a “head count” on the number of enterprises to be privatized (i.e., “divest at least x number of enterprises”). This put the focus on privatizing some rather insignificant enterprises rather than on addressing problems in the small number of state-owned enterprises (SOEs) that had major significance for fiscal sustainability or for the efficient provision of services critical to the private sector; (2) some specified a specific date for a significant privatization (e.g., of a private power project or components of Royal Jordanian airlines). In practice, the government can prepare a company for privatization with all due diligence only to find that at the time of offering it for sale, the market conditions have become adverse and it would simply be imprudent to go forward with the transaction. This is precisely what happened with some privatization cases in Jordan that were subject to such conditions.

Having said that, the overall conclusion of this evaluation is that, with the exception of the 1999 EFF, the design of IMF structural conditionality was generally reasonable. It focused mostly on the fiscal area, it was clearly linked to the achievement of broader program objectives, and the overall number of conditions was kept small. Indeed, the main criticism of this conditionality is the same as that of the overall structural content of the programs—namely that it was too slow in putting emphasis on underlying core reforms to improve the structure of public expenditures.

The findings from the 1994 conditionality review are reported in Schadler and others (1995) and Schadler (1995). On the 2000–02 conditionality review, see IMF (2000 and 2002a). An “Interim Guidance Note” on streamlining structural conditionality was issued to IMF staff in September 2000, ahead of the formal review that took place during 2001–02.

For example, the staff’s 1996 analysis of the prospects for medium-term viability injected greater realism by concluding that Jordan’s medium-term viability remained “questionable” because substantial financing gaps remained in each year of the projection period—implying the need for further rescheduling. It also concluded that the debt ratios were on a “sustainable” downward path and could be reduced to a level consistent with sustained high rates of economic growth—without, however, defining what those levels might be. (See Appendix V in Maciejewski and Mansur, 1996).

IMF staff did undertake analyses of fiscal sustainability issues in the context of Article IV consultations (e.g., the 1995 and 1996 consultations) but not in a manner that analyzed trade-offs between alternative adjustment paths as an input to program design.

See, for example, the paper “Assessing Sustainability” (IMF, 2002b) discussed by the Executive Board in June 2002, which formed the basis for a new framework for assessing countries’ external and public debt.

A summary of the objectives, strategies, and key policies presented in the initial requests for Jordan’s six IMF arrangements is contained in Annex 4.

The most pronounced revisions occurred in 1998 and affected national accounts and fiscal data for 1996 and 1997. This led to a misreporting episode with the IMF (see Box 3.1).

Background papers prepared for the 1995 and 1998 Article IV consultations contained analyses of the sources of growth using a standard “growth accounting” framework but subsequent discussions of growth targets in program documents do not seem to have drawn on the analysis in any substantive way. The 2004 Article IV also contained an analysis of progress on structural transformation and attempted to estimate whether productivity growth had improved in practice. The results suggested that there were some signs of a pickup in productivity growth in recent years—although it is not possible to draw strong conclusions on the basis of observations for only a few years.

The staff’s 2002 fiscal strategy paper in effect criticized the 1999 EFF for relying on the assumption of large unspecified expenditure cuts and proposed an approach based on addressing the underlying fiscal rigidities through specific policy measures that were costed in the medium-term baseline.

Part of the fiscal expansion reflected the introduction of the authorities’ Plan for Social and Economic Transformation (PSET), including investment in social services and basic infrastructure, with an additional cost of about 4 percent of GDP a year to be financed by increased grant financing.

The IMF’s advice on exchange rate policy is also discussed in the context of IMF surveillance in Chapter 4.

For example, the staff report accompanying the request for the 1989 SBA stated that “The government is committed to continue to follow a flexible exchange rate policy and to ensure, as a minimum, the maintenance of the real effective rate.” If pursued in any operational sense—which it does not appear to have been—such an approach would have represented a major potential inconsistency in program design since it would have been hard to reconcile with the objective of rapid reduction in inflation.

Following the move from direct credit controls to indirect monetary control, the limit was set on the NDA of the Central Bank of Jordan (CBJ) from 1995.

Some commentators have suggested the use of an interest rate rule as a trigger for consultation, but for countries with an explicit inflation-targeting framework for monetary policy (see Blejer and others, 2002).

To continue the “belt and braces” analogy, if the tailor has only imprecise information about the size of the customer, applying a “whichever is tighter” rule would probably not yield a good fit for the trousers.

In contrast, a “before-grants” PC could send an inappropriately pessimistic signal in the case of a temporary downturn.

    Other Resources Citing This Publication