Information about Western Hemisphere Hemisferio Occidental
Journal Issue
Share
Article

Panama: Selected Issues and Statistical Appendix

Author(s):
International Monetary Fund
Published Date:
January 2006
Share
  • ShareShare
Information about Western Hemisphere Hemisferio Occidental
Show Summary Details

II. Redesigning Panama’s Fiscal Responsibility Law7

A. Introduction

23. Panama adopted a fiscal responsibility law (FRL) in 2002, which included limits on fiscal deficits and public debt, whose ultimate goal was to contain public debt over the medium term. Its impact in 2003 and 2004 was disappointing. Public debt increased as a result of failure to contain fiscal deficits within the legal limit. The government that took office in September 2004 suspended the FRL until end-2005 while embarking on a medium-term fiscal consolidation that would bring the fiscal deficit targets of the FRL within reach. The new authorities found it unfeasible to meet the deficit limit in 2004 both under a comprehensive and a narrower definition of the nonfinancial public sector (NFPS; see below); they viewed a narrower NFPS coverage as appropriate for the purpose of the FRL, but deemed it still not feasible to meet the deficit ceiling under this definition; and there were ambiguities in monitoring compliance.

24. This paper examines the main features of Panama’s FRL and suggests areas for possible design improvements. The main characteristics of Panama’s FRL are described in section B. The fiscal rules embedded in the FRL are compared in Section C to the features of a “model fiscal rule” presented in the literature. This comparison shows that there is room for improvement in several areas, in particular by clarifying definitions and coverage, and establishing institutional procedures to increase transparency and social control. Section D presents some country experience with FRLs, and explores possible avenues for enhancing the chances of future success for Panama’s FRLs by drawing on Brazil’s and other countries’ experiences. Conclusions are presented in section E.

B. Main Features of Panama’s FRL

25. The FRL approved by the National Assembly in May 2002 reflected agreements reached during a process of consultation with employers, unions and civil society in 2002 (the “National Dialogue”) on: 1) the use of the assets of the Development Trust Fund (Fondo Fiduciario, constituted with the proceeds of privatization); and 2) public debt policy. Chapter 2 of the law (Economic Recovery and Fiscal Responsibility Law, or “Law 20”) contained provisions aimed at reducing public debt levels over the medium term, including a limit on the deficit of the NFPS. Under its Article 11, items 1 and 2, public sector net debt was targeted to decline to 50 percent of GDP over 15 years; in addition, the fiscal deficit was capped at 2 percent of nominal GDP in any given year. Until the targeted debt reduction was achieved, a transitory regime of authorized borrowing would apply (item 3). If the annual growth of real GDP was above 1½ percent, gross debt could not grow in absolute amounts by more than the minimum resulting from two alternative formulas: a) applying to the gross debt stock a percentage increase equal to 80 percent of the nominal GDP growth rate; or b) calculating an absolute amount equal to 2 percent of GDP (the deficit limit). If the growth rate of real GDP was less than 1½ percent, there was no explicit debt limit, but the overarching 2 percent ceiling on the deficit applied. For purposes of the law, nominal GDP of a given year was supposed to be calculated on the basis of the 12 months ended June of that year. Owing to data limitations, in practice GDP on a calendar year basis was used, which left unresolved the question of how to monitor compliance before year’s end.

26. Using historical GDP growth averages, the FRL rules implied a declining net debt-to-GDP ratio over the medium term. Net debt targeted by the FRL was defined as gross debt minus the assets of the Trust Fund and Brady bond collateral. At the time Law 20 was passed, net debt amounted to approximately 70 percent of GDP at end-2001; thus, Law 20 targeted a decline in net debt by about 20 percentage points. A large upward revision in the GDP series implemented during 2002 resulted in a marked lowering of debt-to-GDP ratios. Net debt was reestimated at about 58 percent of GDP in 2002 (69 percent for total gross debt minus 11 percent of GDP in assets). As it turned out, the FRL was ineffective in preventing fiscal slippage. The deficit of the NFPS on a cash basis reached 3.8 percent of GDP in 2003 and an estimated 3.3 percent in 2004; net debt increased to 61 percent of GDP at end-2004. The deviation from the annual debt ceiling was particularly large in 2004, owing in part to the prefinancing in November 2004 of the projected deficit for 2005. The prefinancing, though a sensible strategy given market conditions, would have been precluded, had the FRL not been suspended, showing a limitation to the original design of the FRL.

Ceilings under the Fiscal Responsibility Law, 2002–04
Est.
200220032004 1/
(In percent of GDP)
Nonfinancial public sector deficit
Ceiling2.02.02.0
Actual2.73.83.3
(In millions of U.S. dollars)
Increase in total public debt, gross 2/
Ceiling 3/260324502
Ceiling 4/245257276
Actual1201401,208
Memorandum items:
(In percent)
GDP growth
Real2.24.36.0
Nominal3.94.87.2
(In percent of GDP)
Total public debt, gross69.467.371.6
Net public debt 5/58.355.660.9
Sources: Office of the Comptroller General; Ministry of Economy and Finance; and Fund staff estimates and projections.

The ceilings for 2004 are notional, since the law was suspended in the last quarter of 2004.

Change in external and domestic debt. The gross debt is not consolidated, as it includes central government debt held by the social security agency.

Constraint 1: If real GDP growth is higher than 1.5 percent, the percentage growth in total public debt should not exceed 80 percent of the growth in nominal GDP.

Constraint 2: If real GDP growth is higher than 1.5 percent, the dollar increase in total public debt should not exceed 2 percent of GDP.

Total public debt (gross) minus Fiduciary Fund assets and collateral for public debt (Brady bonds). Does not take into account government deposits.

Sources: Office of the Comptroller General; Ministry of Economy and Finance; and Fund staff estimates and projections.

The ceilings for 2004 are notional, since the law was suspended in the last quarter of 2004.

Change in external and domestic debt. The gross debt is not consolidated, as it includes central government debt held by the social security agency.

Constraint 1: If real GDP growth is higher than 1.5 percent, the percentage growth in total public debt should not exceed 80 percent of the growth in nominal GDP.

Constraint 2: If real GDP growth is higher than 1.5 percent, the dollar increase in total public debt should not exceed 2 percent of GDP.

Total public debt (gross) minus Fiduciary Fund assets and collateral for public debt (Brady bonds). Does not take into account government deposits.

C. How Well Does the Design of Panama’s FRL Measure up to an “Ideal” Rule ?

27. The characteristics listed below have been identified in the literature as the main features of the design of a model fiscal policy rule.8

  • A model fiscal rule should be well defined as to the institutional coverage and the indicator to be constrained. Panama’s FRL lacked clarity in these respects. The FRL was not explicit about the coverage of the public sector, in particular leaving it open whether the Panama Canal Authority (PCA) should be included or not. Although the PCA is a state-owned enterprise, it is autonomous, and its budget is not part of the public sector budget that is approved by the National Assembly. The deficit ceiling was interpreted under the previous government as including the PCA’s large surpluses, which lowered the monitored fiscal deficit.9 The new authorities were of the opposite view, because by the Constitution the PCA has the right to manage its assets, and does not finance the central government deficit. They decided to exclude the PCA from the fiscal accounts monitored for policy purposes (see the related discussion of adequacy below). There was also a lack of clarity regarding the relevant fiscal indicator. Most fiscal accounts in Panama are on a cash basis, and the FRL did not explicitly specify whether the 2 percent deficit ceiling applied to the balance measured on a cash or accrual basis. A cash-based ceiling would provide an unwelcome incentive to build up arrears. A working group set up at the Contraloria to work out the practical details of implementing the FRL ruled in 2003 that an approximation to the accrual deficit was the appropriate benchmark, but this interpretation did not have a clear legal status.
  • The literature stresses the importance of an effective link between the fiscal rule and its proximate goal. The objective of Panama’s FRL was to reduce the public debt burden to a sustainable level. This would call in principle for imposing a constraint on the budget balance as a whole for the widest definition of the public sector. While the PCA’s surpluses have not financed central government deficits, looking forward, the prospective widening of the Panama Canal (at an estimated cost of 30–35 percent of GDP) would put the PCA in large deficits for a number of years. The substantial borrowing involved would increase fiscal vulnerability, through a possible government guarantee or a reduction in the payment of dividends to the government under a hypothetical worst-case scenario. This raises the issue of the appropriate treatment of the PCA in a revised FRL.10 The adequacy of the debt ceiling under the suspended FRL is also open to discussion, as the target under the suspended FRL (50 percent) may leave the country too vulnerable to shocks.11
  • A fiscal rule should be supported by transparency in government operations, including fiscal accounts. This calls for enhanced clarity in the reporting of Panama’s fiscal position and better accounting and statistical standards. Panama’s FRL did not spell out the starting year of application, although the authorities noted in their economic report on 2002 (page 21) that the deficit for 2002 was within the limit set under Law 20. In 2003, the first full year of applicability of the FRL, the need for improvement in fiscal transparency was illustrated by the opaque discussion of the fiscal outcome. Data published by the authorities showed a fiscal deficit of 1.9 percent of GDP on an estimated accrual basis, below the legal limit. However, based on revised and more complete data, the deficit appeared later to have been closer to 3 percent of GDP (about 4 percent, excluding the PCA), well above limit. Owing in part to limitations in the coverage of the budget expenditure system (SIAFPA), there was no established methodology to estimate the deficit of the NFPS on an accrual basis; and the claim that the deficit was below limit was based on provisional data.
  • A model fiscal rule should be enforceable. The FRL seems not to have been complied with in 2003, but there are no consequences of noncompliance in Panama’s FRL, as no sanctions are mentioned, whether financial or judicial, nor any corrective mechanisms triggered.
  • A fiscal rule should be consistent internally and with other macroeconomic policies. In particular, Panama’s full dollarization and the absence of a lender of last resort impose a de facto limitation of prudent domestic bank financing of the fiscal deficit to maintain adequate liquidity. The experience in 2004 seems to indicate that, although the deficit ceiling was breached, a loophole allowed the authorities to bypass the legal limit on borrowing. As bond issuance planned under the budget assumed the 2 percent deficit ceiling would be respected, there was an increased use of short-term bank credit lines from the Panama National Bank (BNP) to finance the deficit as it evolved, which led to a sharp decline in BNP’s net foreign assets. This credit line was subsequently transformed to a long-term loan, which, by reducing BNP’s liquidity and capacity to cushion shocks, was inconsistent with prudent macroprudential management.
  • A fiscal rule should be simple and easy to understand, so that it will be accepted by the public and will be effective for communicating the government’s fiscal policy intentions. Although the medium-term debt objective may be relatively easy to communicate, the annual debt and deficit limits. with references to the growth of both nominal and real GDP, appear complicated.
  • A fiscal rule should allow for flexibility to accommodate exogenous shocks. Although falling short of introducing a cyclically adjusted balance concept, Panama’s FRL sought to address this concern to a modest extent by allowing a relaxation of the debt growth ceiling in periods of slow growth. Alternative ways to allow for flexibility would be to create room for maneuver by targeting a deficit below the legal ceiling when fiscal stimulus is not needed; or allowing by some mechanism a “rainy day reserve” to be built up during an upswing, to be spent to smooth public spending during a slump, subject to well-designed rules governing savings and withdrawals. However, providing for flexibility runs the risk of foregoing simplicity, and applying fiscal rules that are not readily understood and accepted. Moreover, there is also a risk that flexibility could be excessive and undermine the FRL’s effectiveness. Since this risk may be higher during the run-up to elections, Panama’s FRL would need to contain provisions for the conduct of fiscal policy during the political transition, such as quarterly ceilings.12

D. Lessons from Other Emerging Market Countries

28. Over the last few years, FRLs have been introduced in a number of emerging market economies, including in Latin America (notably Argentina, Brazil, Ecuador, Peru) and also in Turkey, the European Union, and other countries. The aim was to provide for a comprehensive legal framework and institutional arrangements to improve fiscal policy outcomes. Assessments of experience with FRLs in emerging market economies are generally inconclusive, with Brazil’s FRL considered the only clear-cut success story in the Western Hemisphere. The mixed experience appears to indicate that policy rules by themselves do not automatically bring about fiscal sustainability and macroeconomic stability. Political economy considerations are deemed to play a crucial role in the chances of success of FRLs, which need to be understood and supported by the public and policymakers.

29. While FRLs have in common that they seek to limit discretion in the conduct of fiscal policy, cross-country comparisons show that there are wide variations in fiscal rules embedded in FRLs. These variations are welcome to the extent that the design of FRLs is tailored to country-specific characteristics, including the extent of decentralization. FRLs differ in scope, content and operational modalities across countries. They put numerical limits on fiscal deficits, public debt or borrowing, with targets defined mainly for the central government (as in Ecuador) or a larger definition of the public sector (as in Argentina, Brazil and Peru). Some FRLs are enshrined in the Constitution, provide for institutional mechanisms of budget execution and reporting, contain escape clauses for exceptional circumstances, and provide for sanctions for noncompliance. Panama’s FRL was a lower-level legislation, relied exclusively on numerical rules, making no procedural provisions, nor providing for escape clauses or sanctions.

30. Assessments of FRLs conducted at the IMF, and in a recent series of essays that analyze various aspects of the experience with FRLs in emerging markets,13 contain lessons which could be useful in redesigning Panama’s FRL.

  • Social agreement. Fiscal rules in emerging market economies need the support of the public to be effective.14 Brazil’s experience illustrates this principle, where the effectiveness of the FRL was enhanced because the fiscal law formalized practices that had already been tested and well understood for some time. As noted earlier, Panama’s FRL reflected agreements reached during a process of consultation. However, broad social support would require that the definition of the deficit and the coverage of the public sector subject to legal limits, and their rationale should be spelled out in the law in order to avoid ambiguities.
  • Credibility. Fiscal rules may be strengthened by structural reforms. In Panama’s context, this would include fiscal and social security reform to underpin the fiscal targets, and good governance to promote transparency and accountability. Moreover, the experience of other countries seems to show that FRLs should not be introduced during periods when countries have to implement a strong initial fiscal consolidation, when the risk of failure is larger.15 The Panamanian authorities have decided to take the time to build the political consensus to implement needed structural reforms, in particular the politically sensitive social security reform, before the FRL would be reinstated in modified form by end-2005. The credibility of this strategy would depend on the implementation of a fiscal reform adopted in early 2005 and a reform of social security during 2005.
  • Numerical rules and procedures. The experience with FRLs seems to indicate that numerical rules work better when supplemented with procedural rules that foster transparency and accountability. Brazil’s FRL successfully combined numerical indicators and institutional procedures. It established annual and intra-annual targets for key fiscal variables, with reviews every two months to assess developments. By law, every four months the finance minister must testify before Congress whether the budget is in line with the targets. Moreover, to protect the budget against rent-seeking, a specific law is required for every new tax expenditure, which together with stringent requirements on the publication of fiscal data, contributes to fiscal transparency. A reformed FRL for Panama could consider similar procedural provisions.
  • Pragmatic practices. Introducing changes to the FRL should be made difficult, to enhance the credibility of the law. In some countries this may call for embedding fiscal rules in the Constitution. However, whatever the legal status of the FRL, pragmatic fiscal practices may be very effective if the spirit of the law is accepted by the public, with buy-in from the policy makers.16 The principle of simplicity seems to rule out deficit limits based on a cyclically adjusted balance; moreover, economic conditions are often volatile and measurement of potential output may be unreliable. A revised FRL for Panama should seek some simpler mechanism for increasing the primary surplus during an economic upswing, and allowing room for automatic stabilizers in a downturn.
  • Public expenditure management systems. They should be sufficiently advanced to help implement the rules. It would be important for Panama to improve the coverage of the SIAFPA to overcome present fiscal data limitations, allow policymakers to have comprehensive and timely information on fiscal developments, and facilitate a quick response to slippages. Also, there is a need to implement a migration to the 2001 GFS methodology to allow a reliable monitoring of the deficit on an accrual basis.
  • Sanctions. Institutional sanctions, for example, suspension of eligibility for elected office, rather than judicial penalties are deemed the most effective ways of ensuring accountability, as they appear easier to enforce.17

E. Conclusions

31. The 2002 FRL has been ineffective in preventing fiscal slippage, and thus the new government suspended the law’s application until end-2005 and will proceed to revise it. While this strategy appropriately allows time to implement a strong initial fiscal consolidation, the credibility of a revised FRL will depend on the implementation of fiscal and social security reform starting in 2005. The medium-term fiscal targets to be set by the revised FRL should be ambitious enough to markedly reduce fiscal vulnerability, and they should be explained to the public to gain wide acceptance. The design of the revised FRL should avoid ambiguities and be widely understood. In this regard, country experience with FRLs and a review of the desired features highlight the need to clarify the targeted measure of the deficit and the coverage of the public sector; establish institutional procedures to increase transparency and accountability; and improve the financial information management system with the aim of eventually adopting the 2001 GFS methodology for accrual-based fiscal accounting.

7Prepared by Eric Verreydt (WHD).
8See “Fiscal Policy Rules,” by George Kopits and Steven Symansky, IMF Occasional Paper No. 162, 1998.
9See “Informe Economico Anual 2002”, Ministry of Economy and Finance, pp. 21–22.
10See Chapter IV for a discussion of this issue.
11As indicated above, the FRL under Law 20 targeted a debt reduction of about 20 percentage points between 2002 and 2017, which, using the revised GDP series, would implicitly lead to a debt ratio below 40 percent of GDP.
12Parliamentary and presidential elections are held around mid-year and the incumbent administration assumes office in September, leaving little time to take corrective measures, if needed.
13“Rules-Based Fiscal Policy in Emerging Markets: Background, Analysis and Prospects,” edited by George Kopits, 2004.
14In his opening remarks on a review of the book cited in the preceding footnote, Deputy Managing Director Carstens observed that, “Fiscal rules can be thought of as a social agreement that fiscal discipline is a critical element of a country’s good governance.”
15For example, Argentina adopted a FRL in 1999, when the economy was in recession and fiscal policy had been loosened in the run up to elections.
16In Brazil’s case, fiscal adjustment efforts were under way when the FRL was adopted in May 2000, and there was a consensus on the need for further fiscal consolidation.
17In Latin America, Brazil’s and Ecuador’s FRLs provide for penal sanctions against public officials, but the lack of automaticity of penal sanctions may undermine their potential effectiveness.

Other Resources Citing This Publication