88612 JUNE 2014 • Number 146 A Framework to Assess the Fiscal Risks of Public Bodies: Jamaica Rohan Longmore, Marta Riveira Cazorla, and Marijn Verhoeven This note outlines a framework for assessing the extent of fiscal risks inherent in the operations of public bodies (PBs), highlighting eight applications in Jamaica. Scenario analysis and stress testing are two components used to determine the combined effects of company- and sector-specific risk factors and macroeconomic shocks on selected PBs. The framework results can help identify the contributions of PBs, as well as determine the PBs’ required budget allocations. The framework also provides an opportunity to determine the potential impact of fiscal risks on a country’s debt dynamics.1 Why Is Assessing the Fiscal Risks of Public These issues raise questions on the adequacy of fiscal Bodies Important for Jamaica? analysis in Jamaica and the extent to which this gap exposes the government to fiscal risks, which, for purposes of this Jamaica has had limited success in its various attempts at fis- note, are defined as shocks that could lead to a deviation from cal consolidation due to structural weaknesses in public fi- budget estimates.4 There is also no framework in place to cap- nancial management policies and processes. The relatively ture and quantify the fiscal risk for PBs. In many instances, high debt level (139 percent of gross domestic product [GDP] the failure to quantify, disclose, and prepare for such risks has at end 2013) has been one of the major outcomes of this frag- resulted in additional government obligations, larger public mented process. In fact, it is widely recognized that the high debt, and, occasionally, refinancing difficulties for the state debt is related to chronic public sector deficits, weak budget coverage, and the fiscal burden of a large number of weakly budget. As a result, fiscal outturns often differ substantially regulated PBs. from budget.5 While there has been significant emphasis on introduc- Fiscal risk assessment is particularly relevant in countries ing systems and frameworks to better streamline the opera- (such as Jamaica) where the fiscal budget is a 12-month cycle tions of central government, reform of the PB segment of (limited long-term forecast) and where cash accounting (or the public sector has been lagging. PBs’ adherence to exist- modified cash accounting) is used instead of accrual account- ing legislation remains a major challenge. Entities continue ing—and where the treatment of contingencies is not clearly to be cited for breaches, including procurement, weak fi- captured.6 As events like the global economic crisis and the nancials, and negligence on the part of fiduciaries. Various euro zone turmoil have shown, fiscal risks affect both devel- committees of Parliament are frustrated because informa- oped and developing countries, with a tendency to become tion presented on PB activities is not sufficient to allow particularly more threatening in countries/companies with proper assessment of fiscal conditions.2 Additionally, plans limited scope for both maneuvering and accessing financing, to have government sign off on performance targets with such as Jamaica. management of each PB to ensure alignment with core The absence of a framework to capture fiscal risk of PBs is functions have not been realized.3 often interpreted as a function of the relative importance as- 1 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise cribed to such risks, limited capacity to develop and maintain The performance of the PB sector has improved in recent a framework to capture and analyze information, or a deliber- years, but risks remain. PBs recorded a deficit of 0.6 percent ate attempt to obscure fiscal exposures due to policy deci- of GDP in fiscal year (FY) 2011/12, representing a significant sions. Examples of the latter could be fiscal risks related to improvement from the average deficit of close to 2 percent of implicit contingent liabilities from banks deemed too big to GDP over the preceding four years (figure 1). This outturn fail, or guarantees that disguise subsidies to selected entities. reflected, among other things, the divestment of several loss- Some governments may also strategically opt not to disclose making entities over this period. all implicit obligations as a way of minimizing moral hazard Net transfers to government from the SFPBs have re- and, by extension, the government’s exposure. This stems mained positive. When comparing the level of transfers from from the fact that by extending explicit financial protection and to government across the 65 monitored SFPBs, it was to PBs, through guarantees or other forms of explicit contin- found that these are highly concentrated in a small number of gent liabilities, the government encourages stakeholders to PBs, with only six SFPBs capturing 80 percent of the govern- take on more risks. ment transfers, and the aggregated contribution of only seven This note outlines a framework for assessing the fiscal PBs representing 85 percent of the total transfers to govern- risks of PBs in Jamaica and then applies it to the eight PBs ment from these 65 SFPBs. with the largest perceived fiscal risk. The results can be used to help determine the macroeconomic significance of selected risks for fiscal sustainability. Box 1. Is the Underperformance of PBs Linked to Negative The PB Sector in Jamaica Incentives? The PB sector, comprising statutory entities and authorities Many of the PBs in Jamaica underperform relative to their as well as government-owned limited liability companies, budget, and, for the most part, depend on government sup- port through subsidies, waivers, and transfers. CAP, for ex- plays an important role in Jamaica’s economy in terms of rev- ample, is a perennial loss-making entity. Prior to the decision enues, expenditures, and employment. The PB sector crosses in 2013 to insulate the budget from CAP’s operations, which 13 sectors and includes a wide range of activities such as de- was also included as one of the structural benchmarks in velopmental, regulatory, social, and commercial. In quite a the IMF program, expenditure outpaced income by almost 2 number of cases, PBs perform quasi-fiscal activities (QFAs), to 1. The entity was running losses of 0.6 percent of GDP, which, when coupled with a net debt equivalent to approxi- that is, public functions on behalf of the government, includ- mately 2.5 percent of GDP, highlight significant government ing regulatory functions and service delivery at subsidized exposure. NWC and JUTC are in similar positions. They are tariffs (for example, for public transport). Many entities incur both loss-making entities with a combined net debt position losses and are heavily indebted and therefore could represent of 3.5 percent of GDP. In the cases of NWC and JUTC, the a call on the budget. This raises questions about the viability losses in part represent uncompensated QFAs, because and the extent of fiscal risks inherent in their operations, as user tariffs for water and bus transport do not cover the cost of investment and operation. CAP, on the other hand, func- well as whether support from the government effectively tions only in commercial markets. compensates PBs for the cost of their QFAs, or rather creates An assessment of the accounts of all three companies perverse performance incentives (box 1). suggests that current operations are such that in the ab- As of August 2013, the number of PBs registered with sence of a commitment to substantial restructuring, the re- the Ministry of Finance and Planning (MOF&P) stood at 195, sidual obligation that would accrue to the government if one were allowed to fail would be quite significant. Notwithstand- of which 90 are characterized as self-financing (SFPBs),7 while ing this fact, as well as the existence of a comprehensive the other 105 entities are included in the central government legislative framework to improve performance, it appears budget and financial reporting. SFPBs are entities that have that not enough is being done to avoid this scenario. The some authority to collect revenue, borrow, and spend outside government has not developed a method to assess QFA of the central government budget, but in most cases with compensation paid to PBs. This creates uncertainty about MOF&P and/or a line ministry’s approval. While these enti- the nature of PB losses—whether they are due to underper- formance of the PB or represent mandated but uncompen- ties are expected to be independent of central government fi- sated QFAs. Enforcement of existing performance standards nances, in practice, the government of Jamaica provides guar- is also weak, the government rarely enforces sanctions antees and implicitly assumes the liabilities of PBs. MOF&P against board members of weak performing institutions. closely monitors and reports on the activities of 65 SFPBs Value for money and performance audits in the public sec- (with gross assets equivalent to 74.6 percent of GDP and a tor are equally rare, which does not bode well for account- ability and transparency. The combination of these factors staff complement of 10 percent of the total public sector creates perverse incentives for maintaining the status quo. workforce). Of these 65 entities, 17 SFPBs are subjected to Source: Authors’ compilation. greater scrutiny given their size and potential for fiscal risk. 2 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise Figure 1. Evolution of Public Sector Balances (as a percent of GDP) perceived risk according to various criteria, which included size (assets, liabilities, and net assets), losses (net income), net 0 1 2 3 4 /8 /9 /1 /1 /1 /1 /1 07 debt (total liabilities minus current assets), and net transfers 08 09 10 11 12 13 1 20 20 20 20 20 20 20 from the government. Entities with the greatest number of ap- -1 pearances in the top rankings of each criterion were selected for further assessment. Applying these risk criteria, the top -3 eight PBs selected for this study were: Clarendon Aluminum Partners (CAP), Jamaica Urban Transit Company (JUTC), Na- tional Housing Trust (NHT), National Road Operating and -5 Constructing Company (NROOC), National Water Commis- sion (NWC), Port Authority Jamaica (PAJ), Petrojam Limited -7 (PJAM), and Urban Development Corporation (UDC). Most of the selected entities were also among those highlighted in -9 the International Monetary Fund’s (IMF) program for special budget balance monitoring as well as entities that will play some role in the -11 public entities balance Government of Jamaica Global Logistics Hub Initiative. public sector balance Although performance of the individual entities varies, -13 the selected eight PBs together account for 2 percent of GDP (measured as total earnings before interest, taxes, deprecia- Source: MOF&P; IMF. tion and amortization). Of the eight PBs, three recorded loss- es in FY2011/12 (CAP, JUTC, and NWC). Not surprisingly, Framework for Assessing Fiscal Risks with the exception of NHT, with its large pool of undistrib- This note builds on the risk framework for state-owned enter- uted refunds, most are significantly indebted, averaging above prises implemented by Indonesia’s Ministry of Finance and 2.5 percent of GDP. The operations of CAP, NWC, JUTC, explained in Verhoeven et al. (2008).8 The framework focuses and NROOC have been beset by chronic losses and accumu- on nonfinancial PBs that represent significant fiscal risks. lated deficit from previous years, which has eroded the capital First, a baseline outlook is constructed based on historical bases of these companies. This indicates that that a recapital- data. The baseline represents the expected outcome of nor- ization effort or a substantial restructuring would be needed mal operations absent of exogenous shocks. This step includes to reduce their substantial dependence on central govern- making assumptions about general macroeconomic condi- ment support, in contrast with the Extended Fund Facility tions, both domestic and international including, among oth- (EFF) conditionality of close to zero balance for selected ers, prices, interest and exchange rates, and GDP growth.9 PBs.10 Notably, transfers from central government to support Assumptions are also made regarding sectoral and PB- these entities amounted to J$15.1 billion (1 percent of GDP) specific factors, including changes in regulations governing in FY2013/14. the sector. Three scenarios (baseline, optimistic, and pessi- Results of Implementing the Fiscal Risk mistic) are assessed to demonstrate the impact of unforeseen Framework developments on the operations of the selected entities. These may be macroeconomic in nature, reflect trends in the The fiscal risk analysis uses firm-level data and benefited from sector where the PB operates, or may be specific to the entity. inputs from the MOF&P as well as selected PBs. The analysis Scenario analysis is a useful tool for fiscal risk analysis be- requires detailed operational insight into the determinants of cause it highlights both risks and opportunities. The next PB revenues, expenditures and balance sheet items, and relies step is a stress test analysis to show how macroeconomic on available information from published PB accounts, annual risks may affect aggregate financial conditions for the PBs. reports, audit reports, and information in the public domain. The key variables typically used to evaluate the financial Information on debt profile and extra ordinary transactions, health and fiscal risks of PBs include net contribution of the where otherwise not available, were obtained through inter- PB to the budget, financing need of the PB, net debt, and off- views of officials from the selected entities. The analysis cov- balance sheet liabilities. ered FY2010/11 through FY2015/16.10 Establishing a baseline Selection and Description of Key PBs The key macroeconomic assumptions used to set the baseline To implement the framework, the PBs representing the largest scenario for this exercise are in line with the macroeconomic source of fiscal risks were selected. In this regard, the 17 more framework of the IMF EFF program and assume a return to closely monitored SFPBs were listed in descending order of moderate GDP growth (averaging 1.4 percent over the next 3 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise three years) and continued significant inflation (averaging 9.7 outstanding payments (mostly to central government), which percent). Additional assumptions largely consistent with the has served to provide a financial buffer, but at the same time IMF World Economic Outlook projections include a moderate raised PBs’ obligations. Increases in such outstanding pay- depreciation of the exchange rate and mostly stable interest ments, as well as depreciation, are the main components of rates and oil prices. Under baseline assumptions, the operat- the adjustment from accrual to cash operational balance. It is ing balance on an accrual basis of the key eight PBs deterio- imprudent to assume that PBs will continue to see the share rates in FY2013/14, to then improve over the medium term of their bills that remain outstanding at the end of the year by about 0.7 percentage point of GDP in FY2015/16 relative increase. As a result, the aggregate operating balance on a cash to FY2012/13. Over the same period, the overall financial po- basis for the eight PBs is set to improve only by about half of sition of the PBs remains broadly unchanged—at 0.3–0.4 per- the increase in the accrual operational balance (.5 of a per- cent of GDP in FY2012/13 and FY2015/16—after a steep centage point of GDP) between FY2012/13 and FY2015/16. decline to -1.5 percent of GDP in FY2013/14. The financial Lowered capital expenditure helps limit the deterioration of performance under baseline assumptions is calculated for the aggregate overall balance by FY2014/15 and reduces the each of the eight selected PBs separately, and then aggregated. needed room to return to positive overall balance by Table 1 presents the aggregation of the financial performance FY2015/16. As a result of ongoing efforts to contain adminis- for the eight entities. The diverging pattern of the operational trative expenditure and enhance operational efficiency in and financial outlook for this set of PBs reflects various trends, PBs, an improved outlook for the aggregate operational bal- which partly offset each other. ance can be achieved without a large increase in capital expen- The difference in outlook for operational and overall bal- diture. For the eight PBs, gross capital expenditures are pro- ances of the eight PBs is mainly attributable to the larger net jected to decline by about .25 of a percentage point of GDP in transfers of PBs to the government. As part of its agreement FY2015/16 relative to FY2012/13. with the IMF, the government has legislated larger transfers At the entity level, the framework used in this study as- from NHT out of its substantial pool of financial resources.12 sumed that CAP’s activity would continue to break even in As of FY2012/13, NHT holds equity of some 8.5 percent of terms of sales cost. Because this would require CAP’s opera- GDP, most of it due to the accumulation of nonrefundable tional losses to stay about the same, government transfers employers’ contributions to the NHT. As a result, net trans- were assumed to continue as in the FY2013/14 budget. fers from the government decline by about .5 of a percentage However, the recent decision not to privatize as planned by point of GDP in FY2013/14. Another key driver for the less end-2013, but to insulate the budget from CAP activities, favorable outlook of overall balances is the rapid increase in will negatively impact its activities relative to the baseline considered at the time the model was prepared. With regard Table 1. Baseline Scenarios for the Aggregate of Eight Selected to the operations of NHT, while remaining very profitable, Public Bodies the first year with larger transfers to the government is esti- Forecast (% of GDP) mated to cause a deterioration to a negative overall balance Performance in FY2013/14. However, NHT will eventually integrate into 2012/13 2013/14 2014/15 2015/16 indicators its activity the larger transfers to government and return to a 1. Operating positive overall balance in subsequent years. PAJ is projected 2.8 2.2 3.1 3.5 balance (accrual) to register operating losses in FY2013/14 and a negative Adjustment overall balance up to FY2014/15, assuming the authorities 1.0 1.0 0.9 0.8 accrual to cash accept responsibility for the dredging of the port (estimated 2. Operating to take 18 months). But eventually PAJ is likely to return to 3.2 3.2 3.9 4.3 balance (cash) profitability, as well as to a positive overall balance by 3. Net transfers FY2015/16, after privatization of the Kingston Container from central -1.3 -1.3 -1.5 -1.5 Terminal (KCT). CAP, JUTC, NROCC, and NWC are pro- government jected to perform at best with a negative, but close to zero, 4. Gross capital overall balance, which would further increase the net debt expenditure on their balance sheets. UDC’s future performance under 3.4 3.4 2.9 2.4 (including inventories) the baseline scenario assumes a very conservative operation going forward, which will avoid the overall balance falling to 5. Overall balance a negative figure. -1.5 -1.5 -0.5 0.4 The PBs whose operations signal very significant levels (-=financing need; 5=2+3-4) of financial risk are NROCC, CAP and NWC—these three Sources: MOF&P; authors’ estimates. have more debt than assets and increasing levels of debt pro- 4 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise jected under the baseline assumptions. A full disclosure of nancial investments and benefit from higher domestic and fiscal risks would include significant risks from this select foreign rates. group of PBs. The financial position of the selected PBs respond differ- Scenario analysis ently to the various shocks considered. For CAP, the analysis An important consideration in scenario analysis is the choice considers changes in operational performance (sales volume of indicators used to assess fiscal risks due to unforeseen and operational efficiency) and borrowing rates. The larger events or shocks. The present analysis considers three indica- (smaller) financing need in the pessimistic (optimistic) sce- tors: net flows to the central government, overall balance (or nario may lead to larger (smaller) transfers from the govern- net financing need) of the PB, and change in PBs’ net debt. ment to CAP. These indicators identify fiscal risks as they emerge through: The scenario assumptions for JUTC relate to the increase (i) an immediate impact on government finances (net flows to of passenger fares, rehabilitation of the deteriorating bus fleet the government); (ii) a reduced scope for sustainable central so that assets can be used productively, and key cost compo- government deficits as PBs increase their borrowing (PB over- nents (fuel prices in U.S. dollars and the exchange rate). In the all balance/financing need); and (iii) an enhanced risk of fu- pessimistic scenario (with higher fuel prices, less favorable ture calls on government assistance to address rising PB in- exchange rate, higher interest rates, and bus fares remaining debtedness, especially when operational revenue lags behind unchanged), JUTC would have very substantial borrowing debt service obligations (PB net debt). In the scenario analysis, requirements in the medium term, about J$3.5 billion or the latter indicator for net debt is related to operational reve- 0.20 percent of GDP in FY2015/16. The possibility of oil nue to provide a sense of the revenues for debt service require- prices rising sharply over the medium term is highly likely ments that the PB is expected to generate. Specifically, the given rising geopolitical tensions in many oil-producing states. analysis looks at the ratio of operational revenue over net It is also not impractical to assume an overshooting of the ex- debt; if this ratio goes up (down), the ability to service the change rate adjustment under the IMF program. Under more debt from operational revenues is presumed to increase (de- favorable assumptions (with lower fuel prices, more favorable cline). In other words, fiscal risk increases as the ratio of opera- exchange rate, lower interest rates, and higher fares), JUTC tional revenues over net debt goes down. would have significant surpluses. Box 2 presents the findings Conducting scenario analysis involves trade-offs re- from scenario analysis for JUTC. garding how changes in operational conditions are translat- The scenarios for NWC focus on capital expenditure ed into balance sheet items. This determines which fiscal (lower in the optimistic scenario), regulated water prices, bor- risk indicator would show the impact, but does not affect rowing rates, and exchange rates. In all scenarios, NWC will the findings of the fiscal risk analysis. For example, one can have a substantial financing need in the medium term. With assume that a shock that unfavorably impacts the opera- transfers from government declining as planned, this will tional balance is reflected in a deterioration of current as- likely entail increased borrowing, which is expected to in- sets, increased debt, or smaller net payments to the govern- crease debt ratios for FY2014/15 and FY2015/16. NWC’s ment budget. This choice is sometimes difficult to foresee, balance sheet, as measured by operational revenue over net unless well-defined policies direct how government trans- debt, will improve only in the optimistic scenario. fers respond to unforeseen circumstances.13 Fortunately, PAJ’s financial bottom line is likely to be strongly influ- the indicators of fiscal risk capture the impact consistently, enced by plans to privatize KCT and developments in trade but not necessarily with the same indicator. In addition to and cruise travel. Under all scenarios, operational revenues as the stated example above, a sharp reduction in current as- a share of net debt would decline as revenues come under sets would be reflected in an increase in net debt; increased pressure from a global decline in business volume (pessimis- debt would show as a lower overall balance/larger financing tic scenario) or revenues from the KCT end due to comple- need as well as increased net debt, while increased pay- tion of planned privatization (in the baseline and optimistic ments from the government would be captured in net scenario). PAJ will need to borrow substantial resources in transfers to the government.14 FY2013/14–15/16 to move forward with plans to dredge the The scenarios have been designed to reflect macroeco- harbor, which will allow Jamaica to profit from servicing larg- nomic assumptions as well as entity-specific potential events er ships associated with the completion of the improvements to illustrate key fiscal risks in the pessimistic scenario as well to the Panama Canal.15 Surpluses in the overall balance will as scope for opportunities in the optimistic scenario. It should return in FY2015/16 under all scenarios. The projected activ- be noted that shocks that hurt the bottom line of some enti- ity points to increasing debt ratios, where by FY2015/16, PAJ ties may have a favorable impact on other entities. For exam- would almost have as much debt as assets. ple, while most PBs see their operational losses increase as a Petrojam is in a very different situation compared to the result of increasing interest rates, some have considerable fi- other PBs. It provides substantial transfers to the government 5 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise Box 2. Example of Scenario Analysis: Jamaica Urban Transport Company (JUTC) a. Net flows to the central government b. Operational revenue/net debta - 2,000 1,000 2012/13 2013/14 2014/15 2015/16 - 2,100 800 - 2,200 J$ millions 600 percent - 2,300 400 - 2,400 200 - 2,500 - 2,600 0 2012/13 2013/14 2014/15 2015/16 c. Net financing needb baseline pessimistic optimistic 4,000 3,000 Sources: MOF&P; authors’ estimates. a. Net debt equals liabilities minus current assets. 2,000 b. Net financing need of the PB equals the overall deficit (or minus overall balance); ↑/↓ indicates optimistic/pessimistic scenario. 1,000 J$ millions Notes: Assumptions: (i) baseline: bus fares increase by 27 percent (zone 1) 0 and remain unchanged (concessional) over FY13/14–15/16; ↑ fare increases are 35 percent (zone 1) and 25 percent (concessional); ↓ fares remain -1,000 unchanged; (ii) baseline: number of buses in operation remains unchanged as problem maintenance improves; ↑ rehabilitation of the fleet raises the number -2,000 of buses in operation by 1 percent by FY15/16; ↓ bus maintenance does not improve and the operational fleet declines by 14 percent by FY15/16; (iii) -3,000 baseline: fuel costs are in line with projections for world oil prices (flat in US$ terms); ↑ same; ↓ fuel prices increase reflecting a rise in world oil prices by -4,000 US$20 per barrel from FY13/14; (iv) baseline: the exchange rate depreciates by 33 percent between FY12/13 and FY15/16; ↑ same; ↓ in FY13/14, the 2012/13 2013/14 2014/15 2015/16 exchange rate depreciations by an additional 20 percent. as it passes on special consumption tax receipts and is not a The aggregate impact of the shocks on the eight selected net debtor like some of the other PBs (its current assets exceed PBs is calculated by year, and then the effects are added over liabilities). Over the medium term, Petrojam’s situation is ex- FY2013/14–15/16. Petrojam is the only PB out of the select- pected to develop favorably, with improvements in net trans- ed eight that benefits from these shocks—with positive net as- fers to the government and net assets as well as in the overall sets (and negative net debt) denominated for the most part in balance. Even if world oil prices decline, capital expenditure U.S. dollars, Petrojam gains from exchange rate depreciation increases and sales growth falls, Petrojam would continue to and increases in interest rates. Petrojam’s bottom line also strengthen its financial position. benefits from higher oil prices. The cumulative outcome of Finally, UDC’s failure to fully insure all its properties the stress tests over FY2013/14–15/16 without the favorable against natural disasters would have a significant effect on its impact on Petrojam is also calculated. financial position in the event of a natural disaster. If 25 per- Significant net debt holdings in U.S. dollars by CAP, cent of its uninsured property is affected, losses could amount NROCC, NWC, and PAJ contribute to a large fiscal risk asso- to about J$2.5 billion (0.17 percent of GDP). ciated with the exchange rate. A depreciation of 20 percent in Stress tests FY2013/14 would increase net debt by 1.7 percent of GDP The stress test analysis focuses on assessing the accumulated for the eight PBs by FY2015/16 and increase their financing impact on the eight selected PBs from shocks on key macro- need (decline in the overall balance) by 0.6 percent of GDP economic variables. The analysis considers an exchange rate over the medium term. As arrangements for covering addi- depreciation of 20 percent in 2013/14, an increase of tional losses from the budget are unclear, the direct impact on US$20 per barrel in world oil prices, and increases in do- the budget is small and represents mostly changes in profit tax mestic and international interest rates of 6 and 2 percentage payments. But, increased strain on vulnerable PBs may shift points, respectively. much of the additional financing need to the budget. There is 6 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise a risk that this could exceed the cumulative effect on the eight delivered, will have significant negative implications for PBs, as Petrojam will see its overall balance increase by 0.4 per- JUTC, NWC, and NROCC. Results here show that, given the cent of GDP. The government may end up with much of the relative size and importance of the entities considered, meet- responsibility for covering the financing need of the remain- ing the objectives for the overall sector as outlined in the IMF- ing seven PBs, amounting to 1 percent of GDP over supported EFF requires careful management of fiscal risks as FY2013/14–14/15, while not receiving substantial addition- well as continued restructuring efforts. al payments from Petrojam. In the absence of a comprehensive fiscal risk monitoring Interest rate and petroleum price shocks have a smaller framework, it is likely that potential sources of risks will be but still considerable impact on PBs’ financial health and fis- overlooked. Adopting and implementing a similar framework cal risks. An increase in domestic and foreign interest rates to the one proposed here, which uses an integrated approach (by 6 percent and 2 percent, respectively) would increase net to identify, disclose, mitigate, and manage fiscal risks present- debt by about 1 percent and financing need by .5 percent of ed by PBs, will help Jamaica better manage contingent liabili- GDP over the medium term. The impact is mostly through ties and streamline its operations. The management and dis- the effect on debt service payments. As for petroleum prices, closure of fiscal risks in Jamaica will need to be facilitated by a the beneficial impact of higher oil prices on Petrojam offsets strong policy framework in which the legal and administra- about two-thirds of the negative impact on the other seven tive arrangement clearly establishes responsibilities and rela- PBs. With limited access to the windfall gains of Petrojam, the tionships among public and private actors, including trans- government still faces significant fiscal risks, as net debt and parent compensation of QFAs from the government budget. financing requirements of the other seven PBs increase by Acknowledgments over .5 percent of GDP over the medium term. The plausible and modest events explored in the stress The authors would like to thank Allan Dizioli, Pedro L. Ro- tests have an added effect of 2 percent of GDP over the medi- driguez, Eric Le Borgne, Auguste Tano Kouame, Francisco um term. This may seem modest, but given present tight fis- Galrao Carneiro, and Sona Varma for helpful comments. cal conditions in Jamaica, such risks would be difficult to Special thanks to the Public Enterprise Division of Jamai- manage if they were to materialize. Also, the impact of less ca’s MOF&P and representatives from the selected entities likely but more severe macroeconomic risks, such as associat- for providing data and invaluable insights into their opera- ed with a global or regional economic crisis, would be much tions. The Auditor General Department also provided use- larger—and much more difficult to deal with. Moreover, this ful insights. represents only the impact of 8 out of 195 PBs—and although the PBs were selected based on their perceived fiscal risk, oth- About the Authors er PBs (such as the Road Maintenance Fund) may also present Rohan Longmore and Marta Riveira Cazorla are Economist and substantial additional fiscal risk. Finally, the stress tests do not Consultant, respectively, in the Poverty Reduction and Econom- include the risks from sector- and PB-specific sources of risk ic Management (PREM) Network of the World Bank’s Latin (for example, related to capital investment programs, regula- America and Caribbean Region. Marijn Verhoeven is Lead Econ- tory price setting, natural disasters, and developments in in- omist in the Public Sector Governance Unit of the World Bank. ternational trade and cruise ship tourism), which were illus- trated in the scenario analysis. Notes Conclusion and Recommendations 1. This note is based on a paper that will be published later this year in the World Bank Working Paper Series. Evidence suggests that the impacts of shocks, whether institu- 2. Less than 50 percent of SFPBs submit financial reports tion or sector specific, or triggered by macroeconomic realign- within the six-month window allowed. The Auditor Gener- ment, pose significant risks to fiscal sustainability. For exam- al’s Office has also stressed the need for PBs to strengthen ple, a shock to the exchange rate could lead to a 2 percent their reporting. increase in the debt stock over the medium term, which for a 3. As noted in the government of Jamaica’s Accountability country like Jamaica is very significant, given existing high Framework for Senior Executive Officers, assessing PB perfor- debt levels. mance has been problematic mainly due to: the governance Significant fiscal risks are embedded in several PBs’ oper- framework of the PBMA Act being less complete than that of ations and should be closely monitored. Specifically, CAP, the Executive Agencies; PBs not promptly fulfilling their re- JUTC, NROCC, and NWC present the greatest risk to the porting responsibilities with the Public Enterprise Division at government’s performance target. Larger than programmed MOF&P; the poor quality of PBs’ corporate plans; and the un- movements in interest and (increased) exchange rates (faster structured method used to objectively analyze performance, rate of depreciation), with no changes in the price of services as well as the absence of performance agreements for staff. 7 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise 4. See Polackova Brixi and Schick (2002) for a more detailed 12. The programmed transfer from the NHT of J$11.4 bil- definition of fiscal risks. lion (0.77 percent of GDP) annually is for a period of four 5. See Jamaica Public Bodies: Estimates of Revenue and Expendi- fiscal years through FY 2016/17 and has been established as ture for the year ending March 2014, http://www.mof.gov.jm/ part of the IMF EFF program. ped/publicdocs. 13. This exists for select cases in Jamaica. For example, Petro- 6. Contingencies may be defined by law or contract (explicit jam transfers special consumption taxes it collects to the gov- liabilities), or arise due to popular expectations and political ernment. Therefore, if tax revenues fall in response to, say, pressure (implicit liabilities; Cebotari 2008). lower fuel prices or lower sales volume, this would be reflect- ed in lower transfers by Petrojam. In other cases, decision 7. See Jamaica Public Bodies: Estimates of Revenue and Expendi- making on transfers is more discrete. This includes most oth- ture (various editions), http://www.mof.gov.jm/ped/publicdocs. er tax payments—for example, profit-making PBs typically do 8. The annual budget documentation in Indonesia includes a not transfer corporate taxes, but instead make or receive a section on state-owned enterprises’ fiscal risks (http://www. lump sum net transfer. anggaran.depkeu.go.id/dja/acontent/Financial%20Note%20 14. This includes the response of PBs to unfavorable events by and%20Indonesia%20Budget%20FY%202013-lengkap.pdf). accumulating payables that would reduce their net assets. The projections of key PB items are contingent on current ex- pectation of operations. Therefore, the exercise should be un- 15. The assumption is that dredging is the prerogative of the derstood as ongoing, requiring updates to reflect any new landlord, which suggests that if the project is to move forward, events of economic significance that may have an influence in it will have to be financed through loans to PAJ. the operation of the entities. References 9. Scenario analysis assesses fiscal risks in alternative states of Auditor General’s Department, Jamaica. 2012. “Performance the world. Key to scenario analysis is the development of a con- Audit.” Urban Development Corporation. http://www.japarlia- sistent set of alternative assumptions rather than using the ment.gov.jm/attachments/784_Review%20of%20UDC.pdf. baseline projections (for example, an increase in oil prices can Cebotari, A. 2008. “Contingent Liabilities: Issues and Practice.” worsen profitability, which could increase a company’s financ- IMF Working Paper 08/245, Washington, DC. ing cost and debt rollover risks). Designing scenarios involves Cebotari, A., J. Davis, L. Lusinyan, A. Mati, P. Mauro, M. Petrie, some judgment (normative exercise) on how the entity and and R. Velloso. 2009. “Fiscal Risks: Sources, Disclosure, and government would react under the alternative scenarios. Stress Management.” IMF, Washington, DC. Davies, J., R. Velloso, A. Mati, M. Petrie, and T. Irwin. 2007. “Indo- tests provide a more mechanical estimation (positive exercise) nesia: Statement of Fiscal Risk.” IMF, Washington, DC. of the impact on the entity from an ad hoc change in one spe- IMF. Article IV Reports. 2001–10. http://www.imf.org/external/ cific variable (for example, the exchange rate) and can be used country/JAM/index.htm?type=9998#56 if the same tests are applied to all key PBs to assess the aggregate IMF, World Economic Indicators Database. fiscal impact of a specific event. Stress tests should be based on OECD (Organisation for Economic Co-operation and Develop- changes that have a reasonable probability of occurring. ment). 2011. “Corporate Governance of State-Owned Enter- prises: Change and Reform in OECD Countries since 2005.” 10. On May 1st, 2013, the IMF’s Board approved a US$932 Paris. million program of engagement with the government of Ja- Polackova Brixi, H., A. Schick, eds. 2002. Government at Risk: maica through an Extended Fund Facility (EFF) to support Contingent Liabilities and Fiscal Risks. Washington, DC: World the country’s comprehensive economic reform agenda. The Bank. EFF includes quantitative performance criteria limiting not Polackova Brixi, H., S. Shatalov, and L. Zlaoui. 2000. “Managing Fis- only the overall deficit of the central government, but also the cal Risks in Bulgaria.” World Bank, Washington, DC. overall deficit of the wider public sector. There are also spe- Verhoeven, M., E. Le Borgne, P. Medas, and L. Jones. 2008. “Assess- ing Fiscal Risk from State-Owned Enterprises.” IMF, Washing- cific targets for selected PBs. For more details on the EFF, see ton, DC. http://www.imf.org/external/pubs/ft/scr/2014/cr1485.pdf. Financial Statement, Annual Reports for the eight selected PBs for 11. The fiscal year in Jamaica runs from April 1 to March 31. various years. The Economic Premise note series is intended to summarize good practices and key policy findings on topics related to economic policy. They are produced by the Poverty Reduction and Economic Management (PREM) Network Vice-Presidency of the World Bank. The views expressed here are those of the authors and do not necessarily reflect those of the World Bank. The notes are available at: www.worldbank.org/economicpremise. 8 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise