Costa Rica
Ex Post Evaluation of Exceptional Access Under the 2009 Stand-By Arrangement

This report reviews Costa Rica’s performance under the Stand-By Arrangement. At the time of the global crisis, Costa Rica’s macroeconomic policy framework was constraining policy options. The financial program’s design was consistent with its policies and objectives. Structural reforms during the program period were slow, and fiscal risk increased. The program’s design reflected the challenges and priorities that Costa Rica faced at that time, as well as policy constraints. Program conditionality focused on maintaining sufficient buffers in the form of foreign exchange reserves.

Abstract

This report reviews Costa Rica’s performance under the Stand-By Arrangement. At the time of the global crisis, Costa Rica’s macroeconomic policy framework was constraining policy options. The financial program’s design was consistent with its policies and objectives. Structural reforms during the program period were slow, and fiscal risk increased. The program’s design reflected the challenges and priorities that Costa Rica faced at that time, as well as policy constraints. Program conditionality focused on maintaining sufficient buffers in the form of foreign exchange reserves.

I. Introduction and Summary

1. The Fund approved a 15-month high-access precautionary stand-by arrangement for Costa Rica on April 10, 2009.2 The global financial crisis had led to a drop in both demand for Costa Rica’s exports and direct capital investment inflows, and the exchange rate had come under pressure. Expectations of continued global financial turmoil led the authorities to seek IMF support to protect the economy against the risk of spillovers and bolster confidence in their policies. The program remained on track until its conclusion in July 2010. The authorities did not draw on any Fund resources made available under the program, which suggests that it served its purpose of shielding Costa Rica from external financing risks.

2. This report reviews Costa Rica’s performance under the stand-by arrangement. Following a summary of the economic context in Section II, the factors leading to the authorities’ request are laid out in Section III. Section IV discusses ownership, and assesses the program’s design and conditionality. Section V reviews program financing and assesses adherence to the Fund’s exceptional access policy. Section VI discusses program outcomes and lessons. The authorities’ views are summarized in Annex I.

3. At the time of the global crisis, Costa Rica’s macroeconomic policy framework was constraining policy options. The exchange rate regime was classified by the Fund as a de facto crawling band. Although preparations were underway towards introducing inflation targeting (IT), monetary policy was constrained by the need to defend the exchange rate band and by weaknesses in the central bank’s balance sheet.3

4. Overall, the program’s design was consistent with its immediate objectives and with Fund policies. These objectives—supporting confidence through providing a liquidity buffer, and making room for the authorities’ macroeconomic strategy to work—were focused on the impending risks. The macroeconomic strategy, outlined by the authorities and supported by the Fund, aimed at countering the threat of external financing shortfall through a tighter monetary stance and alleviating the economic slowdown through countercyclical fiscal policy. This strategy was broadly adequate given the constraints to the policy framework.

5. The program’s immediate objectives were met despite some delays in agreed financial sector reforms. Program outcomes suggest that near-term objectives were met: the external risks did not materialize, and the economy rebounded relatively quickly after the global crisis. The submission4 of laws regarding the financial safety net and the bank resolution framework to the legislature was delayed, but took place before the end of the program. Given the nature of the risks facing Costa Rica and the roots of the global crisis, proceeding with these agreed steps was important.

6. More broadly, however, structural reforms during the program period were slow, and fiscal risks increased. The transition toward inflation targeting was not accelerated, despite what appears to have been a good opportunity to remove the exchange rate band during the gradual appreciation of the colón between mid-2009 and late-2010. Similarly, the authorities did not move to consolidate public finances once the recovery took hold, although staff had called for such a consolidation during the 2009 Article IV consultation. These reforms were not critical in the near term and were therefore not part of the program. Given their importance to the authorities’ medium term goals, however, it would have been desirable to take advantage of opportunities that arose during the program period to make progress on both fronts.

7. Costa Rica has tackled reform in the context of a strong system of constitutional checks. The country has a long tradition of democracy5 and strong institutions. But the legislative assembly has a complicated system of operation, based on a consensus-seeking tradition that seeks to protect minority votes. This often makes it difficult to reach broad consensus on needed reforms. At the time of the program, the scope for reform was further limited by elections upcoming in February 2010.

II. Economic Context

8. Costa Rica’s economy was booming in the years prior to the global crisis. A favorable external environment fueled exports, while strong business and consumer confidence and supportive policies stimulated consumption and investment. As a result, the economy grew at a rate of 6.7 percent on average during 2003-07, well above its potential growth rate, and faster than the region average (Figure 1). Unemployment dropped to levels not seen since the early-1990s, and poverty reached record lows: 16.7 percent of households were below the poverty line6 in 2007, down from 20.6 percent in 2002. The current account deficit increased, but was well covered by FDI and foreign exchange reserves were at comfortable levels (Tables 1 and 2).

Figure 1.
Figure 1.

Costa Rica and CAPDR Growth (y/y)1

Citation: IMF Staff Country Reports 2011, 157; 10.5089/9781455273508.002.A001

Source: International Financial Statistics.1/ CAPDR Includes Costa Rica, Dominican Republic, El Salvador, Guatemala, Honduras, Nicaragua and Panama.
Table 1.

Costa Rica: Selected Economic Indicators, 2003-10

article image
Sources: Central Bank of Costa Rica; Ministry of Finance; and Fund staff projections.

Staff report for the third and final review of the Stand-By Arrangement (June 2010).

Combined public sector = central government + central bank + decentralized government entities + public enterprises, excluding Instituto de Electricidad (ICE).

Includes debt by the Instituto de Electricidad (ICE) guaranteed by the government.

Table 2.

Costa Rica: Selected Economic Indicators, 2006-10

In percent of GDP (unless otherwise indicated)

article image
Sources: Central Bank of Costa Rica; Ministry of Finance; and Fund staff projections

CPS: central government + central bank + other public entities ex Instituto de Electricidad.

Short term external debt: public and private sector external debt at remaining maturity, including trade credits.

9. The authorities used the boom to make progress towards stronger economic and social fundamentals. On the fiscal front, administrative measures, together with growth, helped raise tax revenues by 1½ percentage points of GDP during 2005-07. The authorities were therefore able to increase spending in priority areas such as education, housing and minimum pensions for the poor, while keeping the deficit in check despite slow progress in needed reforms to the tax system. Also, the public debt-to-GDP ratio was substantially reduced over these two years—by 9½ percentage points, to 43 percent by end 2007.

10. There was also some progress towards more effective monetary policy. The focus of monetary policy for many years had been to target the real exchange rate. But inflation had remained stuck at double-digits for most of the time. As a result, the Central Bank of Costa Rica (BCCR) intended to gradually move towards inflation targeting (IT), a framework that would be more effective to reduce inflation. The BCCR was taking steps in line with Fund-provided technical assistance to make this shift possible. In particular, the single rate crawling peg was replaced by a crawling band in October 2006 and the band was widened twice in 2007 (January and November).7 Open market operations were streamlined and the BCCR established an overnight facility. Forward looking elements were also being gradually included in the BCCR’s approach to monetary policy. A law to recapitalize the central bank was submitted to the legislature. But the authorities had not set a clear date for implementing IT; nor had they established a clear course for monetary policy in the transition period.

11. When food and fuel prices surged in late 2007, the economy was overheating but the monetary policy response was not effective in addressing inflation risks. As noted in the staff report for the 2007 Article IV consultation, domestic demand was booming, credit to the private sector was growing rapidly8, and supply conditions were tight. The surge in global food and fuel prices added to these domestic pressures and inflation accelerated, rising back to double-digits by end-2007. At the same time, continued capital inflows and expectations of further appreciation—perhaps driven by a perception that the central bank, in implementing steps towards IT, could relax its management of the colón—kept the exchange rate at the band’s floor (Figure 2). Following high levels of foreign exchange intervention9 which further weakened the central bank’s balance sheet, the authorities resorted to a large interest rate cut at the same time as they announced their intention to bring inflation down, to 8 percent in 2008 and 6 percent in 2009. The authorities, according to the 2007 Article IV consultation report, thought their monetary policy objectives were best served in the immediate term by this approach—otherwise, continued intervention and quasi fiscal losses would undermine policy credibility.

Figure 2.
Figure 2.

Costa Rica: Daily Exchange Rates

(Colones per US$)

Citation: IMF Staff Country Reports 2011, 157; 10.5089/9781455273508.002.A001

Source: BCCR

12. Although financial sector reforms had lagged, risks in that sector were contained. The 2008 FSAP Update emphasized the need for additional steps to make the financial system more resilient. In particular, it called for introducing effective, consolidated supervision and improving the framework for crisis management (including through a system-wide deposit insurance scheme) and bank resolution. But the Update also noted that Costa Rica’s banks remained well capitalized and liquid, and nonperforming loans were low and entirely provisioned.

III. Why Did the Authorities Request a Program?

13. The boom came to an end in the second half of 2008 as the global financial crisis intensified. The sustained rise in food and commodity prices through the summer started to reduce demand. And the recession in the US, Costa Rica’s main trading partner and source of FDI, reduced exports and external financing. Exchange rate expectations quickly shifted towards depreciation and the colón moved rapidly to the top of the band (Figure 2). Inflation peaked at 16.3 percent in November 2008, and growth tumbled to 2.7 percent for the year as a whole, from 7.9 percent in 2007. The external current account deficit widened to 9.3 percent of GDP and, for the first time in recent years, FDI fell short of financing that deficit. The central bank raised short term interest rates in an attempt to stem reserve losses and halt the depreciation of the colón, but the exchange rate remained close to the ceiling of the band. The authorities were hesitant to raise interest rates further given the weak economy.

14. Fiscal policy provided some support, without exhausting available fiscal space. Public spending grew faster than revenues in 2008, as the authorities introduced income-support measures to cushion the impact of the price shock on the most vulnerable segments of the population. Still, the combined public sector registered a small surplus in 2008 (0.2 percent of GDP) and public sector debt was further reduced to 36 percent of GDP by year end.

15. Given expectations of continued global turmoil, the authorities asked for precautionary financial support from the Fund to protect against potential spillovers. They saw the decline in external demand as likely to deepen in 2009, along with a sharp drop in FDI inflows. To strengthen the external position, the authorities planned to tighten monetary conditions. To continue paving the way for inflation targeting, they aimed to gradually increase exchange rate flexibility. The fiscal space created in previous years would allow fiscal policy to cushion the impact of tighter monetary policy on economic activity. Financial sector reforms would also help strengthen the system and limit spillovers from the global financial crisis. The authorities’ baseline program was fully-financed. But investor confidence, considered key to the success of this program, would require liquidity buffers sufficient to absorb large but possible balance-of-payments shocks. To secure such confidence, the authorities turned to international financial institutions, including the IMF.

IV. Program Ownership and Design

A. Ownership

16. The government’s proactive approach towards a program with the Fund before conditions deteriorated to crisis levels represents a sign of strong program ownership. When the authorities approached staff to explore modalities for Fund support, they had already prepared the economic policies for 2009 that would form the basis of the program. Fund financing did not require changes in these policies, which were deemed appropriate by staff; it was rather made available to support an already-designed economic plan. There were extensive discussions regarding the fiscal stance, and program negotiations dragged on for about 3 months before an agreement was reached on the appropriate fiscal position during the program. The authorities had also proactively approached other IFIs to secure additional financing for budget and financial sector support (see below). These are important positive features of the program: strong ownership is generally believed to enhance program success.

17. The program duration also responded to ownership concerns. Because of the forthcoming elections in February 2010, the authorities requested a short duration for the program: 15 months. This choice was reasonable because it helped address the uncertainty associated with a change in government, which could have complicated the achievement of program objectives.

18. The authorities’ communications strategy was judicious. They were proactive at keeping the public informed about the risks that Costa Rica was facing. They explained their efforts to avert an external financing shortfall and made public their contacts with international financial institutions to seek support. These steps helped shore up market confidence, as reported by private sector participants.

B. Macroeconomic and Financial Policy Design

19. The program’s design reflected the challenges and priorities that Costa Rica faced at that time, as well as policy constraints. Monetary policy was tightened to address inflation concerns and exchange rate pressures. The external shock and its impact on growth called for measures to boost economic activity. Several considerations precluded the use of the exchange rate as an instrument to cushion the external shock. First, the exchange rate was not perceived to be fundamentally misaligned. Second, an adjustment of the currency band, particularly on the heels of monetary policy developments in 2008, could have damaged confidence and triggered a currency crisis. Finally, the extent of dollarization implied that the banking sector was vulnerable to a significant depreciation. Fiscal stimulus was the alternative option, within the limit allowed by the country’s fiscal space. Given the stress in the global financial environment, policies to strengthen the financial sector and counter contagion risks were also needed.

20. Monetary and exchange rate policy under the program focused on reducing inflation and defending the exchange rate band, while moving gradually to greater exchange rate flexibility. This policy response began already while program negotiations were ongoing. In January 2009, the central bank raised short-term interest rates on its deposit facility by 70-150 bps to halt pressures on international reserves and control underlying inflation. The authorities also raised the rate of crawl of the ceiling of the exchange rate band from 3 to 9 percent per year. The band would thus widen to about 22 percent by end 2009 and 30 percent by end 2010. To foster competition in the foreign exchange market and reduce intermediation spreads, access to the wholesale component of that market was broadened. Over the course of the program, the authorities committed to adjust interest rates as needed to achieve their inflation targets (set at 9 percent +/- 1 percent for 2009, on end-year basis) and maintain orderly conditions in the foreign exchange market.

21. Steps were built into the program to support the long-planned move to inflation targeting. As noted earlier, implementing IT was a policy objective predating the program. It did not become a program objective; but the authorities intended to make progress within the program towards that long-held goal, with the “hope to complete” it by late 2010.10 Increasing exchange rate flexibility was viewed as an important component. To that effect, the government committed to strengthen the interest rates transmission mechanism. As a first step, the central bank planned to introduce a daily liquidity forecasting exercise to estimate the amount of required central bank intervention in the money market. There was also a commitment to integrate the segmented money market. Both were agreed benchmarks for end-June 2009.

22. Fiscal policy was geared to provide needed economic stimulus through higher social and capacity-building expenditure. The authorities wanted to mitigate the impact of the global crisis and monetary policy tightening on growth—in particular on vulnerable segments of the population. They also aimed at sustaining human and physical capital investment to strengthen competitiveness over the medium term. They therefore planned to use the fiscal space created in previous years to increase spending on education, health, security, and labor-intensive infrastructure projects. Given this target and a projected decline in revenues, the program andexternalenvisaged a substantial increase in the fiscal deficit in 2009—to 3.2 percent of GDP for the central government and 4.1 percent of GDP for the combined public sector. The deficit would be primarily financed from domestic sources, with a contingent World Bank loan for budget support available for any remaining shortfall. The fiscal targets were expected to remain at the same level in 2010, but the increase in debt was expected to reverse over the medium term as the output gap closed.

23. Financial sector policies intended to strengthen the supervisory and regulatory frameworks and the financial safety net. Since the global crisis had its roots in the financial sector, it was important to bolster Costa Rica’s financial sector by eliminating key gaps. The government undertook to submit to parliament draft laws to introduce a bank resolution framework (benchmark for end September 200911) and to create a system-wide deposit insurance scheme12 (benchmark for end December 2009). Closer scrutiny of banks’ liquidity position was also a benchmark for end June 2009, and the central bank committed to keep in place exceptional liquidity assistance measures in colones for as long as they were needed.

24. The government also intended to proceed with other previously proposed policy reforms to strengthen the financial framework. In particular, the authorities committed to push for legislative approval of a draft bill, already in parliament, to recapitalize the central bank. This would enhance monetary policy independence, a key feature of successful inflation-targeting regimes. They also planned to get the approval of a bill to establish consolidated supervision of financial conglomerates and to enhance the sanctions regime. These commitments were not binding in the context of the program as they were mentioned in the memorandum of economic and financial policies without specificity, particularly on timing.

C. Conditionality

25. Program conditionality focused on maintaining sufficient buffers in the form of foreign exchange reserves and on ensuring cautious fiscal policies by containing the stimulus. To maintain investors’ confidence and move credibly towards a flexible exchange rate regime, the floor on net international reserves (NIR) was set at 85 percent of short-term debt for 2009. This level also provided some room for intervention in the event of renewed foreign exchange market pressures. On the fiscal side, the program imposed limits (consistent with the planned stimulus) on the cash balances of both the central government and the combined public sector, as well as on debt accumulation.

26. In the absence of large adjustment needs, and given its precautionary and signaling purposes, program design included streamlined structural conditionality. This was in line with the Fund’s new approach to increased streamlining in the context of recent programs. As noted above, the focus was on two major areas: strengthening the financial sector; and advancing preparations for implementing IT. The choice to avoid fiscal benchmarks was justified by the strong fiscal position of Costa Rica at the onset of the crisis. Boosting financial sector resilience was a priority because that sector was one of the main channels through which the external crisis could impact Costa Rica’s extremely open economy. In this context, conditionality was targeted at addressing gaps already emphasized during the 2007 Article IV consultation and FSAP update. Given the authorities’ commitment to adopt IT and the steps already implemented in that respect, the program called for further steps in that direction to strengthen the interest-rate transmission mechanism.

27. Other forms of conditionality, such as prior actions and consultation clauses, were not needed. Fund arrangements may include prior actions when a member is expected to implement corrective measures critical for the successful implementation of the program before an arrangement is approved or a review is completed. Similarly, consultation clauses with Fund staff may be needed as additional safeguards. Given Costa Rica’s strong track record prior to the global downturn none of these conditionality instruments were considered necessary for the successful implementation of the program.

V. Program Financing and Exceptional Access Policy

28. The resources available under the program were calculated to ensure adequate buffers, needed to boost confidence in the authorities’ policy framework. As noted above, given the uncertain global environment and Costa Rica’s strong linkages to the U.S. economy, financial resources under the program were required to increase liquidity buffers and avoid a disorderly balance of payment adjustment should an external shock materialize. To determine the appropriate level of access, staff looked at the resources needed to preserve coverage of short-term liabilities in the event of potentially large but plausible domestic and external liquidity drains.13 Specifically, the proposed access—492.3 million SDRs, equivalent to 300 percent of quota—was calibrated to preserve NIR coverage at more than 90 percent of short-term debt in the event of an additional decline in exports of 8 percent, a halving of projected FDI, or a decline in the assumed roll-over rates from 100 to 80 percent (Table 3). These assumptions seem to have been adequate: they represented a large but plausible downside scenario for Costa Rica’s economy and were largely in line with the adverse scenarios envisaged in the other high access precautionary arrangements in the region. Consequently, high access was appropriate. The need for high access is not surprising, given Costa Rica’s small quota relative to the openness of its economy as measured by the size of capital flows and trade as well as the potential drains on foreign currency liquidity. It is important to note that total access under the program as a percent of gross financing needs was below most other programs approved around the same period (Figure 3).

Table 3.

Costa Rica: Potential Liquidity Drains and Buffers

article image
Source: EBS/09/41, from Fund staff estimates and BCCR data.
Figure 3.
Figure 3.

Selected Countries: Access

(Percent of gross financing needs)

Citation: IMF Staff Country Reports 2011, 157; 10.5089/9781455273508.002.A001

Source: “Review of recent crisis programs.” (SM/09/246)

29. The authorities’ pro-active and early engagement facilitated burden sharing with other international financial institutions (IFIs). With the global environment rapidly deteriorating, the authorities engaged in early consultations with various IFIs, including the Fund. As a result, additional contingent financing from both the World Bank and the IDB was rapidly secured. The World Bank made available a US$ 500 million budget support loan to be drawn should financing of the public debt not be forthcoming from domestic sources, as envisaged under the Fund program. Similarly, the IDB extended a US$ 500 million loan aimed at providing liquidity assistance to banks should bank funding dry out. These additional sources of financing were never drawn and, in June 2009, the authorities even canceled the IDB credit line, as banks showed little interest in accessing this (fairly expensive) funding source. The World Bank credit line was approved by the Legislative Assembly only by the end of the SBA period, and drawn after the expiration of the program.

30. Substantial frontloading and short program duration aimed at underscoring the crisis prevention nature of the precautionary arrangement (Figure 4). The authorities’ program was geared to boost investors’ confidence and mitigate balance of payment risks. In this context, significant frontloading (200 percent of quota or SDR 328.2 million) was meant to provide a twofold signal: (i) the existence of a large and immediately available liquidity cushion against worse-than-expected external shocks, should they materialize; and (ii) the strong international support for the authorities’ economic program, which included a gradual transition to a more flexible exchange rate regime. The relatively short program duration—15 months—was evidence of the assessed temporary and external nature of the shock and of the absence of fundamental imbalances. It was also consistent with the political cycle, given the 2010 elections that would bring about a change in government. Furthermore, in light of abating external risks and good program performance, the frequency of program reviews was reduced from quarterly to biannual at the time of the second program review.

Figure 4.
Figure 4.

Selected Countries: Frontloading

(first disbursement as percent of total access)

Citation: IMF Staff Country Reports 2011, 157; 10.5089/9781455273508.002.A001

Source: “Review of recent crisis programs.” (SM/09/246)

31. A review of the four exceptional access criteria shows that the exceptional access framework was appropriately applied in considering the authorities’ request14:

  1. “The member is experiencing or has the potential to experience exceptional balance of payments pressures on the current account or capital account resulting in a need for Fund financing that cannot be met within the normal limits’” As staff noted, Costa Rica did experience sizable capital outflows during May-October 2008. Moreover, given the degree of severity of the global financial crisis and the depth of the U.S. recession, there was a significant risk that exceptional capital account pressures emerge—for example, from large cuts in credit lines, or spillovers from a loss in confidence in the banking sector, or a larger-than-expected decline in FDI. Therefore the criterion was met.

  2. “A rigorous and systematic analysis indicates that there is a high probability that debt will remain sustainable in the medium term.” This criterion was satisfactorily met: at the time of the authorities’ request, Costa Rica’s level of combined public sector debt was moderate at 36 percent of GDP, and staff projections showed that the debt-to-GDP ratio would remain sustainable under the baseline scenario and under severe but plausible shocks. Prudent fiscal policies prior to the crisis and the absence of any potential contingent government liabilities provided assurances that debt levels would remain sustainable in the medium-term.

  3. “The member has prospects of gaining or regaining access to private capital markets within the timeframe when Fund resources are outstanding.” Although the government and private sector access to private capital markets had worsened as a result of the global crisis, credit to banks and large corporates had not been cut off. Furthermore, staff was of the view that because of the moderate debt level, the good policy track record and the promising economic outlook, Costa Rica had good prospects of regaining full market access at good terms once global financial conditions improved. In fact, throughout the program, the sovereign did not tap the international capital markets as it was able to secure all needed financing from domestic sources. Therefore, the criterion was appropriately applied.

  4. “The policy program of the member provides a reasonably strong prospect of success, including not only the member’s adjustment plan but also its institutional and political capacity to deliver that adjustment.” Costa Rica entered the crisis with strong political and economic institutions, and a track record of generally sound macroeconomic policy implementation. Furthermore, policies supported by the Fund program appropriately responded to the risks faced at the time, and had a huge degree of ownership These factors provided reasonable assurance that program outcomes would be successful.

32. As required under the exceptional access policy, early Board involvement took place in the period leading up to the program request. A Board meeting was held on January 26, 2009 to discuss the need for a stand-by arrangement involving exceptional access. In line with the exceptional access policy, a report assessing the risks to the Fund and the Fund’s liquidity position was provided to the Board. The staff reports and the authorities’ Letter of Intent and Memorandum of Economic and Financial Policies have been published.

VI. Program Outcomes and Overall Assessment

33. The program remained on track as risks declined. The authorities’ policies conformed to agreements under the program. Economic performance against quantitative targets was strong: all quantitative performance criteria were met, often with large margins. This, and reduced uncertainty about the global environment, allowed for a reduction in the frequency of program reviews from quarterly to biannual, which was agreed during the second review.

A. Elements of Success

34. Success of a precautionary program when the contingency does not materialize is difficult to assess; but on balance the program was successful in meeting its stated goals. The external shock against which the program was providing a cushion did not materialize. This is probably in part because global economic conditions started to improve after the approval of the program. But as staff reported to the Board during program reviews, policy makers and market participants widely held the view that buffers put in place by the program also helped to safeguard Costa Rica’s external balance of payments. Indeed, following the program approval, confidence-related outflows of capital appear to have slowed or halted; and exchange rate pressures abated as the exchange rate fluctuated away from—although still very close to—the upper band. It is therefore reasonable to suppose that the program helped to buttress market confidence, a key objective at the time.

35. Costa Rica’s macroeconomic performance compares favorably with that of other countries (Figure 5). The downturn in Costa Rica was comparable to that in other emerging countries in general, and milder than that experienced by other program countries, while the export rebound was stronger. Remarkably, Costa Rica managed to maintain very stable debt levels whereas the broad emerging market universe experienced substantial increases in public debt. More favorable initial conditions played an important role. But the program also helped by ensuring stable conditions under which a growth rebound could take place.

Figure 5.
Figure 5.

Costa Rica: Relative macroeconomic performance, 2007-101/

Citation: IMF Staff Country Reports 2011, 157; 10.5089/9781455273508.002.A001

Source: WEO1/ Non-program Region includes Belize, Honduras, and Panama. All the country-group series are averages within groups.2/ Costa Rica data as of December 2010.

36. The economy rebounded within the lifespan of the program, and inflation and external imbalances remained in check. By late 2009, economic activity had started picking up again, with manufacturing leading the recovery; and business and consumer confidence were turning around. Inflation decelerated sharply in the course of 2009, making room for cautious interest rate cuts mid-year. By early 2010, headline inflation was edging up, but largely owing to administered price increases; core inflation remained subdued, although the recovery was rapidly taking hold. The balance of payments position had strengthened considerably, and the exchange rate was on an appreciation track (Figure 2).

37. The banking sector remained sound during the program (Table 4). The authorities were proactive in supporting the system as the global crisis unfolded, including through measures to boost liquidity and increase the capital of public banks. Maintaining such support under the program has likely helped preserve financial system stability. A modest deterioration in nonperforming loans and provisioning took place in 2009, as the economy weakened. But it reversed in 2010, as growth strengthened. The authorities had to intervene in one financial institution15, but there were no spillovers to the rest of the system.

Table 4.

Costa Rica: Banking Sector Indicators, 2007-2010 1/

In percent

article image
Source: Superintendency of Banks.

Year-end data for 2007-09. Latest available observation for 2010.

B. Challenges Ahead

38. Although the program helped Costa Rica weather the crisis well, better outcomes could have been possible. During the 2009 Article IV consultation (which coincided with the first program review), staff had highlighted the risk of insufficient progress on three fronts—fiscal consolidation as soon as the recovery takes hold; strengthening the monetary policy framework; and addressing residual gaps in financial sector supervision and regulation. When the program ended, these issues remained outstanding. Legislation to strengthen the financial system had not been approved; progress towards inflation targeting was sluggish; and no steps were underway to consolidate the weakened fiscal stance.

39. Looking forward, economic reforms that require legislative changes will need to accommodate Costa Rica’s consensus-based political system. The benchmarks related to legislation on bank resolution and on the safety net (Table 5) were delayed mainly to allow for further consultations. This is in contrast with benchmarks related to the IT framework (daily forecasting of liquidity and the unification of the money market) which were timely met. To date, the financial sector laws remain under discussion in parliament, as is the case for the central bank recapitalization law, a crucial element in the authorities’ plan to move to IT.

Table 5.

Costa Rica: Structural Benchmarks

article image
Source: Costa Rica Staff Reports.

40. Gaps in the monetary policy framework also remained after the program ended. Daily management of monetary policy operations, such as central bank interventions in the foreign exchange market, remained unclear. And efforts to introduce inflation targeting (IT) had stalled. In retrospect, the gradual appreciation of the colón from summer 2009 onwards seems to have been a good opportunity to speed up these efforts, although not all of the elements supporting an IT regime were in place at the time. But the authorities did not seize the opportunity, partly because the imminent political transition was not, in their view, amenable to such a radical policy shift at the time.

41. Costa Rica came out of the program with a much weakened fiscal position, in part due to the nature of the counter-cyclical fiscal policy implemented during the crisis (Box 1). In retrospect, a positive fiscal impulse was appropriate in both 2009 and 2010 given the negative output gap (Table 6). But most expenditure measures implemented under the program were permanent: increases in public sector wages and pensions, and new hires in the education and security sectors. Wage increases were initially planned for a small group of employees (costing below 0.1 percent of GDP) but were extended to other workers, raising the total cost to some 1 percent of GDP. Weaker-than-expected revenues also explain the significant increase in the deficit under the program, as tax revenue projections proved to be optimistic. The fiscal position could have been improved had a comprehensive tax reform been implemented. However, only minor taxes were introduced under the program (including on luxury homes).

Table 6.

Costa Rica: Selected Economic Indicators, 2006-10

In percent of GDP

article image
Sources: Central Bank of Costa Rica; Ministry of Finance; and Fund staff projections

CPS: central government + central bank + other public entities ex Instituto de Electricidad.

Old measure for the structural balance that was used during the program. Staff currently use a different methodology, whose results underlie the sign of the fiscal impulse shown at the bottom of the table.

The underlying output gap is calculated based on a Hodrick-Prescott trend.

Most recent staff estimates of the fiscal impulse (as of March 7th, 2011).

42. Significant uncertainty over the course of the program also contributed to this fiscal outcome. In particular, the timing and speed of the economic recovery, both globally and in Costa Rica, was subject to significant uncertainty. Through late 2009, doubts about the strength of the global recovery persisted, with direct implications for economic prospects in Costa Rica where, up through the second review, the economic outlook remained fragile. On balance, allowing for continued stimulus through the first quarter of 2010 seemed the right way to go, in order not to undermine the recovery. Only by the time of the third and final review had it become reasonably clear that the recovery was taking hold and, by that time, there was limited scope to adjust the program to reduce the fiscal impulse.

43. The deterioration in the fiscal position under the program raises the question whether additional fiscal conditionality was needed. In retrospect, it could have been appropriate to include an adjustor that would link the fiscal balance under the program to the economic cycle. If such an adjustor could have been put in place at the onset of the program, it might have helped contain the fiscal expansion as the economy started recovering. But such measures are hard to calculate, let alone explain to the public. They would be particularly hard to calculate for Costa Rica, a small open economy largely exposed to external shocks and for which comprehensive high-frequency real sector data are not available. Two other approaches could have been explored:

  • Ask the authorities to channel the fiscal stimulus through temporary rather than permanent measures, or to put a ceiling on recurrent spending. Such conditionality may have undermined program ownership and could have been seen by the authorities as unwelcome interference in the details of their fiscal program which, in the authorities’ view, was responding to a long-standing need for strengthening public services.

  • Include tax reform in program conditionality. Staff and the authorities must have expected that the impact of permanent expenditure increases on fiscal sustainability would be addressed through revenue reform in due course. Formalizing this at the onset of the program was not considered, given the strong starting fiscal position and because parliament was already deeply engaged in discussing CAFTA legislation. Had tax reform been made part of the program, either at its onset or later in the course of one of the program reviews, it might have increased the emphasis on the need to address fiscal sustainability over the medium term. However, it probably would not have helped change fiscal outcomes given the upcoming change in government and the slow pace of Costa Rica’s legislative system.

Fiscal Policy Under the Program

The Fund-supported program allowed for an increase in the budget deficit that was not expected to threaten fiscal sustainability. In 2009, expenditure on wages, transfers, and capital investment would be raised significantly to alleviate the impact of the economic slowdown. As a result, the structural balance of the central government would decline by 2.3 percent of GDP. For 2010, the budget deficit was envisaged to remain constant at the 2009 level, implying a small improvement in that structural balance under the agreed macroeconomic framework (Table 6). Over the medium term, with growth resuming, staff expected the authorities would be able to keep public sector debt below 40 percent of GDP but noted that this goal would require the authorities to enhance revenues, perhaps through tax reform, should the increase in spending become permanent.

During the course of the program, the deficit targets were increased. By the time of the first review in August 2009, the economic slowdown was deeper than expected, and there was a sharp contraction in highly-taxed (durable goods) imports. As a result, projected revenues for 2009 were marked down during the first review, and deficit targets were revised upwards to protect government spending (Table 6). During the second review (November 2009), the overall expenditure envelope was maintained at broadly the same level as it was in the first review, with reduced capital expenditure compensating for an expansion in wages and pensions. Deficit targets were maintained constant at the 2009 level for 2010 in both reviews. This implied a small fiscal withdrawal for 2010 under the projected macroeconomic framework. But this withdrawal was cut by half in the first review compared to the negotiated program: the implied change in the structural balance between 2009 and 2010 was 0.2 percent of GDP in the first review, compared to 0.5 percent of GDP in the negotiated program. Also the quarterly pattern agreed during the second review allowed for a positive fiscal impulse in the first quarter of 2010: with the recovery still fragile and uneven across sectors, staff and the authorities agreed that there was still need for supporting domestic demand.

With the recovery stronger than expected by end 2009, fiscal policy turned out ex-post to be looser than planned for the year. At the time of the third review, the estimated fiscal outturn in 2009 was better than expected on the account of (capital and wage) expenditure restraint and a larger-than-projected one-off surplus of the social security fund. Therefore, the deficit targets that had been set for 2010 in the second review implied a slightly expansionary fiscal stance for the year, rather than the withdrawal previously envisaged. But during the third program review the authorities noted that it was not possible to tighten fiscal policy in the second half of 2010 given the political transition (a new government was taking office in May 2010.) The latest estimates put the 2010 central and consolidated government deficit at 5.5 percent of GDP, a full percentage point above expectations at the time of the final review. A weak recovery in tax revenues largely contributed to this outcome, but also increases in current and capital expenditures related to security, education and infrastructure.

The absence of concrete proposals to reduce the deficit raised concerns regarding the medium term fiscal position. The staff report for the third review envisaged an increase in the public debt ratio to 44 percent of GDP by 2015 under a no-policy change scenario. To help reverse this increase in indebtedness, and restore the fiscal space used during the crisis, staff reiterated the advice (previously made during the 2009 Article IV discussions) to start fiscal consolidation in 2011.

C. Key Lessons

44. Costa Rica’s experience with the 2009 precautionary program has demonstrated the following key lessons:

  • Streamlined and well-targeted conditionality can be helpful in programs with short duration and focused immediate objectives. The light conditionality increased ownership and facilitated both program negotiation and implementation.

  • That said, it is difficult to strike the right balance in setting conditionality, given the trade-off it sometimes presents between ensuring ownership and meeting program objectives. For Costa Rica, despite the global crisis, including an element of structural fiscal conditionality may have helped contain the deterioration in the fiscal position although it would have jeopardized program ownership.

  • Any future engagement with Costa Rica needs to take into account two constraints: the consensus-based legislative process and, in the absence of IT implementation, monetary policy’s constrained room for maneuver. The program navigated these waters relatively well, by avoiding conditionality that would require rapid parliamentary action and by easing the monetary policy bias to support the economy as soon as exchange rate pressures abated. Given the lengthy legislative process, it is advisable that the government work with the legislature to approve as soon as possible key laws, now in parliament, that are needed to address gaps in the financial sector and consolidate the fiscal position. Also, to remove constraints on monetary policy, it is important to move forward rapidly with IT implementation. Completing these steps would strengthen the country’s position ahead of the next external shock if (or when) it comes.

  • The nature of stimulus matters. This is an old lesson relearned: stimulus through permanent expenditure increases will, other things equal, undermine the fiscal position over the medium term. Whenever possible, it would be preferable to achieve the same objectives or target the same vulnerable groups through temporary measures. Otherwise, and to the extent possible, early steps to counter permanent stimulus with corrective fiscal measures down the road would be advisable.

  • When possible, changes in conditionality should be pursued as needed to address changes in the macroeconomic environment. Such flexibility may help achieve better outcomes. In the case of Costa Rica’s program, adjusting conditionality during the second review, if it were possible, might have helped contain the deterioration in the fiscal position.

D. Concluding Remarks

45. The Fund’s program with Costa Rica met the country’s needs in the wake of the crisis. It allowed the authorities to respond to the external shock and alleviate its impact on the exchange rate and on real economic activity. For a number of reasons noted above, the program did not readily lend itself to a platform for seeking progress on longer-term reforms. Costa Rica’s financial reforms and adjustments to its monetary policy framework preceded the crisis and progress on these fronts is proceeding, albeit slowly. Speeding this process would be important to prepare the country for the next external shock, should it come soon. It is also now time to tackle fiscal consolidation, which was partly undermined by the stimulus allowed within the program.

Annex I. Authorities’ Views16

46. The authorities agreed that the program had met its immediate objectives. The program’s effectiveness in shoring up confidence had already been felt during the negotiations, as the government and the IFIs were sending reassuring signals about likely successful program negotiations. This view was echoed by private sector participants in Costa Rica. A visit to the country by the IMF Managing Director in November, 2008, in which he publicly stated the IMF’s readiness to support Costa Rica, had also been helpful. In the authorities’ view, high upfront access was a key contributor to the success of the program in shoring up confidence.

47. The central bank noted that it would have been difficult to make faster progress on reforms to the monetary policy framework. Ex-Governor Gutierrez indicated that several factors had made it impossible to remove the exchange rate band when the exchange rate was gradually appreciating in 2009-10 away from the bands. Initially, the recovery was still fragile and any shock to policy would have risked undermining it. By the time the recovery had taken a stronger hold, elections were around the corner and the Governor was keen to avoid policy changes that may become the focal point of the political debate. Subsequently, when Governor Bolanos took office, the window of opportunity had passed and the exchange rate was already at the most appreciated end of the band.

48. The former authorities defended the conduct of fiscal policy under the program. They made a number of points in support of the measures taken at the time. First, there had not been enough “shovel-ready” investment projects to provide the needed stimulus via capital expenditure. Second, the increase in current government spending was required to boost public sector productivity, reduce wage gaps within the public sector, and make needed improvements in public sector services, particularly education and security. They acknowledged that the fiscal position deteriorated more than anticipated, but partly blamed it on overly optimistic tax revenue projections. It had not been possible to push for tax reform under the program because the government had spent all its political capital on the legislative approval of CAFTA in the run-up to the presidential elections. But the authorities expected tax reform to be approved this year as all parties understand its essential contribution to Costa Rica’s fiscal consolidation plans.

2

This was the second in a series of three HAPAs in the region, and the first HAPA after the Fund facilities reform, approved by the Board on March 24th, 2009, formalized these programs. The other two cases were El Salvador (approved on January 16, 2009) and Guatemala (approved on April 22, 2009), for which ex-post evaluations were also prepared.

3

A recapitalization law that would aim at restoring strength to the central bank’s balance sheet was in parliament (and still is, to date).

4

These submissions were structural benchmarks under the program.

5

Costa Rica has had a continuous democratic regime since its brief civil war in 1948.

6

The authorities define the poverty line as the income level under which basic necessities (food and shelter) cannot be met.

7

In mid-2008, the floor of the band was set flat and its width temporarily narrowed, but starting 2009 the width of the band was increased as the ceiling was allowed to crawl at a higher rate.

8

Credit to the private sector grew at annual rates ranging from 30 to 40 percent throughout the year.

9

The central bank purchased US$422 million (2.9 percent of GDP) in the first 4 weeks of January 2008.

10

This was specified in the MEFP, as a possibility, not as a commitment.

11

During the first review, it was agreed to move this deadline to December 2009 to coincide with the timing of the deposit guarantee law because of the complementarities between these two laws.

12

This scheme would be along the lines recommended by the 2008 FSAP Update. Costa Rica does not have a deposit insurance agency, and public banks have an advantage over their private counterparts given the fullguarantee of their liabilities by the state.

13

In this context, empirical evidence suggests that countries with high initial buffers in relation to short-term liabilities fared better during the global financial crisis (see SM/10/116, “How Did Emerging Markets Cope With the Crisis?”).

14

Costa Rica’s request was for exceptional access was assessed under the modified exceptional access criteria (SM/09/69, Sup. 2, 3/24/09), which broadened the first criterion to include cases where the member “has the potential to experience exceptional balance of payments pressures.”

15

The 4th largest cooperative, COOPEMEX, was intervened in February 2010 because it had persistent capital shortfalls. But this was a small institution (less than 1 percent of total banking sector assets) and its problems predated the crisis.

16

The EPE team visited Costa Rica during March 14-15, 2011 and met the current and former authorities, and members of the private sector to seek their views on the program.

Costa Rica: Ex Post Evaluation of Exceptional Access Under the 2009 Stand-By Arrangement
Author: International Monetary Fund