Investment Trends

Over the past decade, total investment in the Caribbean region has been trending down. Following a period of strong growth, total investment (as a share of GDP) started to decline (Figure 1) with the global financial crisis and economic downturn, which negatively impacted the economies in the region. Most countries had a lower level of total investment during 2010 to 2015 relative to 2005 to 2009 (Figures 2 and 3), except for Haiti (in light of the reconstruction following the devastating 2010 earthquake), the Bahamas, Dominica, and Guyana. In the Pacific region, the total investment-to-GDP ratio has remained lower than in the Caribbean but on a broadly steadier path, whereas investment in peer countries in SSA has trended upwards from 2003, before leveling off over the recent period.

View Full Size

Public and Private Investment Trends

(Percent of GDP)

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: IMF Fiscal Affairs Department Investment and Capital Stock Dataset.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Caribbean Countries: Investment, 2010–15

(Median, percent of GDP)

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Sources: Haver Analytics; Penn World Tables; and IMF staff calculations.Note: Data not available for Jamaica.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Caribbean Countries: Investment, 2005–09

(Median, percent of GDP)

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Sources: Haver Analytics; Penn World Tables; and IMF staff calculations.Note: Data not available for Jamaica.

• Download Figure
• Download figure as PowerPoint slide

Private investment, which plays a key role in the region, experienced a large decline. Over the last three decades, private investment has been twice as high as public sector investment and largely above levels reached among comparator groups. But private sector investment in almost all Caribbean countries has been trending down in the last decade, driving the decline in total investment. Only Barbados, Bahamas, and Guyana have managed to scale up private investment during this period. More recently, private sector investment in the Caribbean region picked up again; however, it is still well below its 1999 peak.

Governments in the region only temporarily and partially offset the decline in private investment. Initially, more than half of the Caribbean countries had a marginal increase in public investment, aiming to support growth during the global and domestic downturn. However, from 2008, countries in the region curtailed capital spending in response to deteriorating public finances. Thus, public sector investment in the region decelerated from about 10 percent of GDP in 2005 to about 6½ percent of GDP in 2015, significantly below levels in peer countries in both the Pacific region and Africa. Almost two-thirds of the Caribbean countries scaled down public investment, except Barbados, Grenada, Belize, St. Kitts and Nevis, and St. Lucia.

PPPs are another way for governments to procure public infrastructure. A PPP is a project governed by a long-term contract between a government and a company. The latter makes an investment in an asset and, using that asset and perhaps other assets made available by the government, provides services to the government or the public (Irwin, Mazraani, and Saxena, 2018). The company is usually private² and is typically established specifically for the project. The services are usually ones for which the government has traditionally been responsible, such as those provided by roads, railways, schools, hospitals, prisons, or airports. The government continues to have some responsibility for the quality of the services and to bear some of the attendant risks of providing them. At the end of the contract, control of the asset typically reverts to the government.

While PPPs have been extensively used in some Caribbean countries, they have not significantly contributed to public investment since the global financial crisis.³ Some governments in the region resorted to PPPs for large-scale infrastructure in the first half of the 1990s, notably in the electricity, transport, and water and sanitation sectors (Caribbean Development Bank 2017b; World Bank 2014a). Thus, Caribbean countries have on average the highest PPP capital stock as a share of GDP relative to peers in the Pacific region and in Africa. However, PPP investment in the region has been declining since 2005 to levels below peer countries in SSA. As a result, PPP investment did not offset the declining trend in public investment over the past decade (Figure 4).⁴

View Full Size

Private Partnerships Investment and Capital Stock^{1, 2}

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: IMF Fiscal Affairs Department Investment and Capital Stock Dataset.¹Caribbean average includes Belize, Dominica, Grenada, Haiti, Jamaica, and St. Lucia; Pacific average includes Fiji, Maldives, Samoa, Solomon Islands, Timor-Leste, and Tonga; sub-Saharan Africa average includes Cape Verde, Comoros, Mauritius, São Tomé and Príncipe, Seychelles, and Swaziland.²Total investment = public investment + PPP investment; total capital stock = public capital stock + PPP capital stock.

• Download Figure
• Download figure as PowerPoint slide

Few countries have implemented multiple PPPs. Jamaica has implemented large PPP projects (see the example of the Sangster International Airport in Box 1) and mobilized private investment in infrastructure through PPPs (The Economist Intelligence Unit 2017). Most countries in the region have less experience in negotiating and implementing PPPs. For example, in the airport sector, private participation is limited. Most airports in the Caribbean are owned and operated by the public sector, with only Jamaica and the Bahamas having functioning airport PPPs. Recently completed and ongoing airport expansion projects in Antigua and Barbuda, St. Vincent and the Grenadines, and Guyana are all publicly-funded and operated. Private sector participation in the Caribbean airport sector is challenging because few airports would pass the industry practice of at least one million passengers per year. Beyond PPPs, the Caribbean region has a long history of partnership between governments and the private sector in the ownership and operation of resort hotels. The typical model is a lease agreement for hotel assets owned by the government with an international hotel operator.⁵

Infrastructure Provision

Despite the decline in public investment, the value of the public capital stock per capita in the Caribbean region remains relatively high on average (Figure 5).⁶ Caribbean countries have on average a high public capital stock per capita (adjusted for purchasing power parities) relative to their peers in Sub-Saharan Africa and the Pacific. This masks however significant disparities across Caribbean countries. Trinidad and Tobago outperformed all Caribbean countries, while Haiti lags its peers in the region.

View Full Size

Public Capital Stock per Capita, 2015

(2011 PPP$-adjusted, thousands)

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: IMF Fiscal Affairs Department Investment and Capital Stock Dataset.Note: PPP$-adjusted = purchasing power parity adjusted for US dollars.

• Download Figure
• Download figure as PowerPoint slide

Access to economic infrastructure in the region has improved substantially in recent years, but Caribbean countries lag on social infrastructure (Figure 6). In this regard, the Caribbean countries have on average outperformed their peers in SSA and the Pacific, particularly in roads and in electricity production. However, with respect to social infrastructure (education and public), the region has fallen behind the Pacific region and appears on par with SSA comparators. Access to treated water is almost universal, as in the Pacific region, and to a lesser extent in Africa.

View Full Size

Access to Infrastructure

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: World Bank, World Development Indicators.Note: Units vary to fit scale. Left scale: Public education infrastructure is measured as secondary teachers per 1,000 persons; electricity production per capita as thousands of kWh per person; roads per capita as km per 1,000 persons; and public health infrastructure as hospital beds per 1,000 persons. Right scale: access to treated water is measured as percent of population.

• Download Figure
• Download figure as PowerPoint slide

The quality of existing infrastructure in the Caribbean region has suffered in recent years (Figure 7). Overall, infrastructure quality has somewhat declined in the Caribbean region and is perceived as lower than in SAA comparators on average. A breakdown of the overall index⁷ shows that, in terms of the average quality of their air transportation and ports, Caribbean countries are on par with their African peers but performing better than their peers in the Pacific. Given the importance of tourism in the Caribbean region, the expectation was that the region would outperform its peers with respect to the average quality of both its air transport and ports. Only Barbados, and to a lesser extent Jamaica and Trinidad and Tobago, have a better score than the average in other regions. Haiti is significantly behind on almost all indicators for quality of infrastructure, illustrating its longstanding development challenges, which have been exacerbated by the devastating earthquake in 2010.

View Full Size

Caribbean Region: Infrastructure Quality Indicators

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: World Economic Forum; and IMF staff estimates.Note: Caribbean region average includes Barbados, Haiti, Jamaica, and Trinidad and Tobago.

• Download Figure
• Download figure as PowerPoint slide

Limited Fiscal Space

Most Caribbean countries are constrained by limited fiscal space to increase capital spending. With an average fiscal deficit of about 3 percent of GDP, the fiscal situation in the Caribbean region has been worse than for its peers in the Pacific and African regions (Figure 8). Starting in 2010 though, Caribbean countries have experienced a worsening in their aggregate fiscal balance. The countries in the Caribbean regions also have a higher debt service than their peers in the Pacific and African region, limiting their fiscal space and flexibility, in particular Dominica, Belize, Jamaica, and St. Kitts and Nevis.

View Full Size

Caribbean Countries: Fiscal Situation, 2007–18

(Percent of GDP)

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: IMF, World Economic Outlook database.

• Download Figure
• Download figure as PowerPoint slide

High public debt is also affecting some Caribbean countries’ access to international financial markets. Continued weak fiscal balances have contributed to a sizeable accumulation of public debt in the Caribbean region. The Caribbean economies are much more indebted than the average of their peers in other regions (see Figure 9). Despite decreasing in the mid-2000s, driven in part by restructuring operations and bilateral debt relief (Schipke, Cebotai, and Thacker, 2013), average debt levels remain high at about 75 percent of GDP. There is a considerable diversity in debt burdens across the Caribbean, with some countries with a debt-to-GDP ratio below 60 percent of GDP, whereas others have a debt ratio above 100 percent of GDP, which has led markets to assign high sovereign risk to most countries and thus reduces access to international financing (IMF 2013a).

View Full Size

Nominal Public Capital Stock and Gross Debt

(Percent of GDP)

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: IMF World Economic Outlook database.

• Download Figure
• Download figure as PowerPoint slide

Financing Constraints

Access to long-term financing for infrastructure is hampered by shallow domestic financial markets. Financial sectors in the region are large relative to the size of the regional economy, with total assets averaging 320 percent of GDP, but not geared toward infrastructure investment. The financial sectors in these countries are in general dominated by banks that have shorter lending horizons relative to infrastructure investment (IMF 2013b). In some countries, the insurance sector is large relative to population, but it is not substantially financing local infrastructure investment due to the absence of well-developed local capital market. Some countries have an important credit union sector, which is mainly providing consumer and mortgage loans. And the very large offshore banking sectors is geared toward international clients and does not provide domestic financing.

The Caribbean region has limited access to global capital markets. Most Caribbean countries have limited access to global capital markets, as international investors are reluctant to take on small country exposure, given the risk posed by economic volatility, high indebtedness, and the cost of administering and monitoring relatively small financial transactions (Alleyne and others 2017; IMF 2013b, 2014A). This has resulted in illiquid markets for debt in most countries in the region and, therefore, access to financing for both public and private infrastructure investment has been heavily constrained. While external debt financing is growing in some countries—such as Trinidad and Tobago and Barbados—it remains practically nonexistent in the other countries of the region.

Most Caribbean countries cannot compensate the lack of private financing for infrastructure by tapping into concessional financing. While concessional flows have increased globally, including after the global economic crisis, most Caribbean countries—having reached upper-middle- and high-income status—have become less successful in accessing international development assistance as aid preferences shifted toward low-income countries and countries in post-conflict situations (Bourne 2015). Haiti is the largest recipient of official development aid (ODA) in the Caribbean region, receiving 80 to 90 percent of total aid disbursement to the region. Total bilateral flows to Caribbean countries (excluding Haiti) have declined substantially since 2009, reaching a very low level in recent years (see Figure 10). This decline took place as countries were dealing with the adverse effects of the global recession and were facing dwindling fiscal space for capital investment. With the exception of Barbados and Trinidad and Tobago, which received no ODA, Caribbean countries currently receive ODA ranging between 0.1 percent and 4.9 percent of their gross national income, with Haiti at the upper end, followed by Dominica and St. Kitts and Nevis with 4 percent, and Guyana and Belize with 3.3 and 3.2 percent of gross national income, respectively. Other Caribbean countries are at the medium to low end of the aid spectrum. Overall, Pacific island small states have been more reliant on development assistance than Caribbean countries (IMF 2013b).

View Full Size

Caribbean Countries: Total Bilateral Development Aid

(Millions of US dollars, constant prices)

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: Organisation for Economic Co-operation and Development, International Development Statistics database.

• Download Figure
• Download figure as PowerPoint slide

Small Size Constraints Infrastructure Investment

The small size of most Caribbean countries limits the potential for achieving economies of scale in infrastructure investment. Most Caribbean countries are characterized as small states based on their population size and, as such, these countries face substantial challenges in achieving economies of scale. Lack of economies of scale in providing public goods and services can limit institutional capacity because of fixed costs, raises the average cost of the public sector for small states, and results in the under-provision of some public goods and services (IMF 2013b). This is a challenge for the provision of infrastructure, as there is limited scope to generate sufficient return to cover fixed costs of investment given small market size.

Vulnerability to Natural Disasters and Climate Change

The Caribbean region is exposed to natural disasters. The region has experienced a series of natural disasters that have caused tremendous losses in terms of human lives, productive assets, physical infrastructure, and production, especially in agriculture and tourism (Appendix I). The average annual economic cost of disasters during 1990 to 2014 was equivalent to about 2.5 percent of GDP for the region, six times higher than in other countries and about ½ percentage point of GDP higher than in other small countries (Figure 11) (IMF 2016).

View Full Size

The Carribean Region Susceptibility to Natural Disaster

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: EM-DAT databases.

• Download Figure
• Download figure as PowerPoint slide

Natural disasters in the Caribbean region are not only costlier, but also more frequent than in other countries (Figure 11). The frequency of natural disasters reached an all-time high in the 2000s and the level reached in the first half of 2010s decade is already comparable to earlier decades (1960s and 1970s) and is on its way to reach the levels of the 1980s and 1990s. The frequency of disasters varies significantly within the Caribbean region, with Jamaica and the Bahamas having a more than 20 percent probability of being struck by a hurricane in any given year, and most of the other Caribbean countries having a probability higher than 10 percent.

At the same time, many islands in the Caribbean region are exposed to climate change (Box 2). Climate change is expected to exacerbate the impact of natural disasters, increasing the frequency and force of hurricanes, causing sea level to rise—threatening coastal infrastructure, beaches (crucial for the tourism sector), housing settlements, fresh water supplies, marine life, and coastal fisheries.

Climate change can cause extensive damages to the economy and infrastructure in the Caribbean region. The economic cost of climate change for Caribbean countries is projected at about 8 percent of GDP per year associated with one-meter sea-level rise (SLR) by 2080 (IMF 2016) and would largely impact infrastructure:

• Coastal infrastructure and ecosystems. SLRs and more intensive and frequent tropical storms could affect coastal infrastructure, destroying buildings, roads, airports, hotels, marinas, and other facilities in the tourism sector. Erosion of beaches is also a major source of risk for the tourism-depending Caribbean countries.

• Transport systems. Intense rainfalls can cause flash floods and landslides, which can destroy bridges and road infrastructure—leading to high repair costs.

• Port activities. More intensive tropical storms, combined with SLRs, could also affect port activities, raising shipping costs, reducing trade, and hampering cruise tourism.

• Energy services. SLR-related intensive tropical storms could threaten energy infrastructure in the region. Storms with high winds could damage pipelines, power transmissions, and distribution lines, resulting in power shortages.

Risks associated with natural disasters and climate change may discourage private investors or force governments to become the insurer of last resort (Public-Private Infrastructure Advisory Facility 2016). Natural disasters and climate can adversely impact the physical infrastructure, increase capital and operating expenditures, cause service disruptions and income losses, or increase insurance costs (World Bank 2016a). While typically a PPP contract allocates risks between the public and the private partners, climate risks are generally not considered or allocated to a specific party (Public-Private Infrastructure Advisory Facility 2016). In that context, private companies may be reluctant to invest in large infrastructure projects given significant uncertainties. And if PPPs are executed, the government may become the insurer of last resort to ensure availability and continuity of infrastructure services in case risks arising from natural hazards and climate change materialize.

Gateway Process

A structured “gateway” process with a strong role for the minister of finance is critical for safeguarding public finances against fiscal costs and risks from PPPs.¹ The gateway process should give the minister of finance the authority to stop or suspend a project at any stage of the project cycle, including project appraisal, assessment of PPP appropriateness, tendering, contract closure, and renegotiation (Table 2). This requires that the role of the minister of finance goes beyond endorsing the final decision of a cabinet or a procuring agency. International experience shows that the capacity for a minister of finance to stop a PPP project by checking budget affordability just before contract awarding is very limited, given the political and social pressures that has built up at this stage to deliver the project. Therefore, the minister of finance should be able to check budget affordability at early phases of the project cycle to prevent the implementation of projects that do not provide value if implemented as a PPP or are not affordable in light of current and future budgetary constraints. Also, the minister of finance should have a veto power over contract renegotiation and termination and be part of the monitoring process of PPPs during both the construction and operational phase, to limit fiscal risks generated by challenges in implementing PPP contracts.

This process can integrate the role of a planning or infrastructure ministry, as a decision-making entity with respect to investment planning and project approval. The ministry of planning would ensure that projects are in line with government investment priorities and are well prepared. It is essential that the minister of finance retains a veto power to stop projects that are not affordable or that would not be delivered efficiently through a PPP.

The Caribbean countries generally do not provide their minister of finance with a gateway role. In most Caribbean countries surveyed, the decision on PPPs is taken by the cabinet or a high-level steering committee and the minister of finance does not have the power to stop PPPs that are not affordable (Figure 18).

View Full Size

Decision-making Powers in Public Investment and Public-Private Partnership Process—Practice in the Caribbean Counties

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: IMF regional survey.

• Download Figure
• Download figure as PowerPoint slide

The gateway process should apply to all public investment projects, whether procured traditionally or through a PPP. Projects from all levels of government should be subject to this process. All types of investment projects, regardless of how they are funded and procured, or under which sector they fall, should be covered. This would ensure a level playing field for all investment projects, regardless of the mode of implementation. Such a unified approach for all investment projects that is linked to the budget process ensures that all projects are subject to the same assessment and that all projects are competing over the available resources in a transparent allocation process.

Institutional Framework

Governments have created units that promote the use of PPPs and serve as a center of excellence. International experience points to the usefulness of entrusting a central unit with the overall responsibility for formulating policy and providing guidance on PPPs (European PPP Expertise Centre 2014; World Bank 2017). Many countries involved in PPP contracts have established a PPP unit, staffed by economists, financial analysts, accountants, and lawyers undertaking a wide range of functions. They include (1) advising the government on PPP policy, (2) developing and disseminating knowledge and expertise on PPPs, (3) assessing specific proposals for PPP projects and programs, (4) advising on the preparation of tender documents and on bid evaluation, (5) undertaking legal and economic reviews of specific contracts, (6) evaluating the results and operational risks of running PPP contracts, and (7) oversight and control of PPP projects. Typically, units report to the respective ministry of finance, as they are either located within the ministry of finance or as separate executive agency of that ministry (European PPP Expertise Centre 2014). In some countries with large PPP programs, line ministries have also developed their own centers of expertise, working in collaboration with the central PPP unit (Organisation for Economic Co-operation and Development 2010).

The responsibilities of PPP units should be distinguished from the control function performed by the ministry of finance (Organisation for Economic Co-operation and Development 2010). Conflicts of interest may arise when the same public entity promotes PPPs and is tasked with the responsibility and empowered to stop or suspend project that undermine the sustainability of public finances. The same entity that advises a contracting agency in identifying and/or selecting a project should not approve or monitor it. In many countries the PPP unit advises the contracting agencies in various technical aspects related to PPP projects, and given lack of experience and expertise with in the rest of the administration, the PPP unit is often also involved in the approval and monitoring process. However, a clear separation between the advisory function of a PPP unit and the control function of the budget department or the Treasury department or the risk management unit should be ensured, so that these departments never participate in the contract negotiation and award, as the ministry of finance should primarily validate them through its role in the gateway process.

In the Caribbean region, the institutional framework for managing PPPs is limited. Only six countries (Haiti, Jamaica, Saint Lucia, Trinidad, Turks and Caicos) have put in place a PPP unit or a PPP team that regroups staff from different departments (Grenada) that generally sits in their ministry of finance. In most of these countries, the PPP unit has clearly defined roles and responsibilities and intervenes at all key stages of the PPP process. This, however, includes signing-off fiscal risks before project approval in two of these countries (Grenada and Saint Lucia), hence creating a risk of conflict of interest, whereas the PPP promotion and fiscal oversight functions are well segregated in Jamaica.

Caribbean countries should develop an integrated gateway process for all public investment projects, including PPPs. The entity in charge of the control function of PPPs and other infrastructure projects should be clearly identified within the ministry of finance, as almost half of the Caribbean countries do not have an entity designated to enforce the gateway process. This entity should be different from the PPP unit, as for example in Barbados where the budget office is in charge of the control function, and the PPP unit plays a role of center of excellence. Jamaica provides another example of segregation between the promotion function and the control function to prevent conflicts of interest. The PPP unit is housed within the Development Bank of Jamaica, and a separate PPP unit in charge of fiscal oversight is established within the ministry of finance (Caribbean Development Bank 2017a).

In addition, Caribbean countries should better integrate PPP management into PIM. There have been important institutional and regulatory changes in the PPP framework in many countries. Advanced countries tend to better integrate traditionally procured public investment and PPPs to pursue PPPs for improving efficiency and not to circumvent fiscal constraints (World Bank 2017). For example, France, South Africa, and the United Kingdom have recently integrated public investment and PPP management functions into their Treasury department. To support all large public investment projects, irrespective of how they are implemented, the mandate of existing PPP units in Caribbean countries should be expanded to turn them into infrastructure units rather than PPP units. In other Caribbean countries, creating such an infrastructure unit might not be advisable, given the small size of the ministry of finance and the limited number of large investment projects and PPPs implemented or planned. Functions related to supporting the development and implementation of large public investment projects, including PPPs, could be provided by the regional infrastructure unit recently created, by expanding the existing initiative of the Caribbean Development Bank (Box 9).

Identifying Fiscal Risks from Public-Private Partnerships

It is important to understand the fiscal risk from its entire portfolio of PPPs. The decision of a new PPP project or the management of fiscal risks of individual projects is based on a project-by-project approach to assess fiscal implications. However, for a government to manage public finances, the risk of the overall PPP portfolio is more relevant. Due to the accumulation of fiscal commitments and the interlinkages of different projects, and the resulting correlations in the projects’ risks, the firm and contingent liabilities of all PPPs must be looked at together.

The IMF and the World Bank have developed the PPP Fiscal Risk Assessment Model (PFRAM) to assess fiscal risks and costs from PPPs. This tool can be used to assess the long-term fiscal implications and fiscal risks (that is, contingent liabilities) of PPP projects. The PFRAM allows to quantify the impact a new project would have on the headline fiscal indicators (that is, government deficit and debt) over the lifetime of the project. It shows how a baseline projection of fiscal outcomes would change if the project was undertaken on top of the ongoing government activities (Box 10). A new version of PFRAM is being developed to analyze a portfolio of PPP projects and consolidate their fiscal implications. It can be used to assess the existing portfolio of PPPs and/or to the impact of a new project or several new projects on the existing portfolio.

Assessing Fiscal Risks from Public-Private Partnerships

The decision on whether to enter into a new PPP should include an ex ante assessment of the long-term fiscal risks of the PPP project under consideration. In traditionally procured projects, the decision on whether the fiscal implications of a project are manageable for the government is made through the budget process, where the allocation of the limited available resources for the current budget year and, through the MTBF, the medium-term is decided. Because the fiscal implications of PPPs typically go far beyond the time horizon covered by the budget and the MTBF, there is no expenditure limit that could guide whether the fiscal implications of a new PPP would be manageable for government finances over the long term. This is also the case for most traditionally procured large-scale infrastructure projects as the time required for their implementation may exceed the horizon of the MTBF.

In the PPP lifecycle, the risk profile of a project may change and therefore, must be updated. The PFRAM can be used to assess the implications of a project at an early stage of the project development, using assumptions regarding the project and contract structure, funding, and financing, and can be refined as the project details are firmed up. It can also be used to assess the fiscal implications of ongoing projects, as for the Barbados Solid Waste Management Facility (Box 11).

In the region, most countries do not have a framework to assess fiscal risks at the key project stages. According to the survey, only five countries (Barbados, Jamaica, St. Lucia, St. Vincent and the Grenadines, Turks and Caicos) require fiscal risk assessment at project selection, tender preparation, contract negotiation and renegotiations, and contract extension. Other countries have either no fiscal risk assessment required (such as Guyana) or only at selected project stages (for example, Haiti).

Once risks for a specific PPP are identified, their fiscal impact can be estimated using alternative evaluation techniques. The quantification of the fiscal impact depends on the type of fiscal cost or risk, and to what extent the analysis incorporated the probability of occurrence or likelihood. Maximum exposure (face value) and expected fiscal exposure (weighted by likelihood) are important measures to fully understand the overall fiscal implications of a PPP project. When likelihood is not considered, or the future event is regarded as certain, fiscal costs and risks are estimated at maximum exposure (face value). This is what is reflected in the results generated by PFRAM. If likelihood of the risks materializing is considered in the analysis, fiscal exposure is estimated as fiscal impact times likelihood. In order to do this, the PFRAM results would have to be adjusted. Table 3 summarizes examples of different methods to quantify fiscal implications of PPPs based on international experience.

Table 3.

Assessing Fiscal Costs and Risks from Public-Private Partnerships

Sources: World Bank 2016c; and IMF staff.

Table 3.

Assessing Fiscal Costs and Risks from Public-Private Partnerships

Type	Example	Fiscal Impact	Likelihood
Fiscal Costs	Gap viability payments Availability payments Output-based payments	Present value of annual payments over the project life	Certain, probability = 1
Fiscal Risks	Guarantees	Scenario analysis: Present value of estimated annual payments (baseline)	Under alternative scenarios for main risk trigger variables (for example, GDP, inflation)
		Probabilistic analysis: Expected value of estimated annual payments	Stochastic simulations modeling changes in risk trigger variables (for example, GDP, inflation)
	Termination payments	Option-valuation techniques: Estimate probability of default Maximum value of termination payment under baseline assumptions	Structure model Contract dependent

Sources: World Bank 2016c; and IMF staff.

View Table

Table 3.

Assessing Fiscal Costs and Risks from Public-Private Partnerships

Type	Example	Fiscal Impact	Likelihood
Fiscal Costs	Gap viability payments Availability payments Output-based payments	Present value of annual payments over the project life	Certain, probability = 1
Fiscal Risks	Guarantees	Scenario analysis: Present value of estimated annual payments (baseline)	Under alternative scenarios for main risk trigger variables (for example, GDP, inflation)
		Probabilistic analysis: Expected value of estimated annual payments	Stochastic simulations modeling changes in risk trigger variables (for example, GDP, inflation)
	Termination payments	Option-valuation techniques: Estimate probability of default Maximum value of termination payment under baseline assumptions	Structure model Contract dependent

Sources: World Bank 2016c; and IMF staff.

Simple and intutitive methodologies for estimating fiscal costs and risks from PPPs are the best option for Caribbean countries. To this end, scenario analysis tends to be a good starting point. Probabilistic and other complex techniques require a significant amount of data on the underlying variables triggering risks (long-term series, assumptions of steady state levels, and so on). As a result, they can be difficult to implement and interpret, reducing the credibility of the analysis for decision making. In practice only a few countries use sophisticated analysis to assess risk exposure to the PPP portfolio. In the Caribbean region, the results provided by PFRAM could be used as a starting point to assess the fiscal impact of PPP projects in a very simple way.

Reporting for Public-Private Partnerships in Small Island Context

International accounting and reporting standards suggest that most PPPs should be considered on governments’ balance sheet. In 2011, the International Public Sector Accounting Standard (IPSAS) 32 set out the accounting requirements for public sector entities as the grantor in service concession agreements. Similarly, the Government Financial Statistics Manual 2014 (GFSM 2014) defines international standards for how to treat PPPs in determining the government deficit and debt (IMF 2014b). For both accounting and statistical treatments, whether a PPP is considered on- or off-balance sheet and therefore how it impacts fiscal headline indicators, depends on whether the PPP-related asset is regarded as public or private. The decision on whether to regard an asset as public or private under IPSAS 32 and GFSM 2014 is based on who controls the asset and on the economic ownership of the asset, respectively (Box 12). While these two concepts differ in the specific criteria that they apply, the decision on whether to account for an asset on the government accounts or not is mostly the same.

Accounting for and recording on PPPs’ on-balance sheet eliminates the bias in favor of PPPs. As discussed, PPPs have often been used when fiscal constraints did not allow to finance infrastructure projects from the budget. A cash budgeting and accounting environment encourages this as the short-term implications of a PPP differ materially from those of traditionally procured infrastructure as the net-cash outflows are postponed and spread over a longer time, often decades. However, the long-term impact on public finances is much the same if a project is implemented through a PPP or through traditional public procurement. When applying IPSAS 32 and GFSM 2014 for accounting and reporting, the impact of PPPs on headline fiscal indicators (that is the deficit and public debt) is very similar to that of traditionally procured projects. Therefore, the “wrong” incentive for undertaking PPPs—just because they allow to push investment off-budget and off-balance sheet—is eliminated.

Reporting and accounting for PPPs is country specific and so is the capacity to avoid the bias in favor of them. IPSAS 32 has been designed for application by entities following accrual accounting. However, many governments do not currently follow accrual accounting standards. Some present fiscal data only on a cash basis and do not have a balance sheet prepared according to any specific standard. Consequently, most governments do not recognize PPPs on their balance sheets or treat investment in PPPs as public investment in fiscal data.

In the Caribbean region, few countries apply international accounting standards. According to the survey, less than 15 percent of the Caribbean countries applied accrual accounting standards. A recent study covered a different sample of 15 countries from the region and came to a similar result (International Federation of Accountants 2017). Countries in the Caribbean who have adopted accruals continue to experience reporting challenges. The first country to adopt accrual accounting, the Cayman Islands in 2004, has difficulties with state-owned enterprises and compliance by line ministries, departments, and agencies. Bermuda adopted accruals in 2010 with a three-year plan for compliance and continues working to improve recording. Barbados started planning for accruals in 2002 and adopted them in 2007, but financial reporting issues continue for pensions, assets, and accounting consolidation. In addition, Aruba and Curaçao are in the process of applying accrual accounting.

Several countries are in the process of or are planning to move toward accrual accounting standards. According to the International Federation of Accountants 2017 study, one-third of the 15 Caribbean countries covered by the survey will remain on cash basis. The remaining countries are planning to adopt or to transition to accrual standards. However, the introduction of accrual accounting comes with significant challenges and takes time. Box 13 gives an overview of the challenges these reforms might pose for some Caribbean countries.

More Caribbean countries apply international accounting standards, but with significant limitations for PPPs. According to the survey, while seven countries have implemented GFSM 2014 in the region, only three of them are producing a government balance sheet. In addition, only three countries report accounting for PPPs in both fiscal deficit and public debt (Barbados, Jamaica, Turks and Caicos) (Figure 20).

View Full Size

Reporting Practice for Public-Private Partnerships in the Caribbean Counties

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: IMF regional survey.

• Download Figure
• Download figure as PowerPoint slide

Intermediate solutions are required to report PPP liabilities in the Caribbean region. It is important that governments take short-term actions to compensate the adverse implications of cash-based accounting and reporting in the context of PPPs. These actions should aim at informing transparently about the implications of PPPs on the fiscal deficit and public debt. The latter can be done in a clear and transparent manner by including PPP-related liabilities in the government debt reporting. Governments can also improve transparency over fiscal implications by including detailed information on all firm and contingent costs of PPPs on future budgets in budget documents and financial reports.

According to the survey, the information on fiscal risks is only available in three countries in the region. Caribbean countries could align with international good practices in other emerging market economies, such as Georgia (Sayegh and others 2017) and the Philippines, which already publish fiscal risk statements that include PPP-related liabilities. While this does not impact the budget deficit and thus does not put the PPP on a level playing field with implementing the project through traditional procurement, it informs about the long-term implications a project is expected to have on public finances.

The information provided through the accounting and reporting systems must be considered when decisions on new projects are made. Good accounting and reporting practices can help reveal the long-term fiscal implications of PPPs. However, the accumulation of fiscal risks from PPPs can only be managed if the expected fiscal implications of new projects are known at the time when the decision on the project is made. This can only be achieved if information on the long-term implications a project would have, irrespective if it is implemented through traditional public procurement or through a PPP, is provided when new projects are identified, prioritized, and selected in the budget process. Once a project has been selected and implementation has started, the government can only manage but no longer avoid fiscal risks.

Finally, auditing of PPPs is generally limited in the Caribbean region. Ex post assessment of PPPs by an independent institution creates incentives for government to check affordability and value for money when preparing a project and help improve transparency. According to the survey, PPPs are generally not required to be audited by the supreme audit institution, and only three countries have had audits of PPP projects.

Mitigating and Managing Fiscal Risks

Governments in Caribbean countries should decide how to mitigate or manage existing fiscal risks. Once the risks are understood and reported, the government needs to decide how to deal with them. Well-structured contracts and an active supervision and enforcement of contract implementation are often the best way to limit and mitigate fiscal implications and risks. However, this requires an experienced team with the capacity to manage such contracts. Most Caribbean countries—due to the size of their economy—would not have a sufficient number of PPPs to gain the necessary experience or the manpower to focus on following PPPs closely throughout their lifetime. Governments generally need external advisors to support them in the preparation and negotiation of the project. However, this comes at a high price, which must be factored into the overall project cost estimates.

To mitigate fiscal risks, governments can limit their PPP exposure. This can take the form of ceilings on the total debt or total investment value that can be accumulated or contracted under PPPs, on the annual expenditure for PPPs, or limits on government support for PPPs, such as guarantees (Box 14). Ceilings or caps, while not a substitute for medium-term planning, can help contain fiscal costs and limit overall government long-term commitments to levels that are fiscally affordable. Only two Caribbean countries (Jamaica, Turks and Caicos) have binding ceilings for the total volume of PPPs.²

To manage fiscal risks from PPPs, governments in the region can transfer risks or share them with the private partner. To realize efficiency gains through a PPP, project risks must be assigned to the party that is best equipped to manage them. For example, the government will typically be in a better position to manage political risks (for example, changes in tax or labor laws, civil unrest or war), whereas the private sector can generally handle construction risks more efficiently. By clearly assigning all identified risks, the public and the private partners can develop strategies for managing the risks under their responsibility (Box 15). This will come at a cost but in case of limited risk management capacity, the cost of keeping and not managing the risk might be even higher. Risks that cannot be mitigated or transferred to another party need to be accommodated by the government (that is, the budget needs to provide room for taking up the cost should a risk materialize) or through insurance whenever possible and affordable. In addition, while the responsibility for and thus the actual management of some risks may be allocated to the private partner, the government should have monitoring mechanisms in place to ensure that the private partner takes on their responsibility and manages the risks actively. Otherwise, in case the private partner neglects their responsibilities, the risks that have been explicitly assigned to the private partner might become implicit risks to the government.

Governments can use the PFRAM risks matrix to identify and manage fiscal risks that are difficult to quantify. In addition to a quantification of the firm and contingent liabilities related to the project and their implication on the government finances, the PFRAM provides a risk matrix. Users can assess risks that might not be easily quantified and obtain guidance for managing fiscal risks from the individual PPP. Some of those risks are explicitly allocated to the government, but many others are implicit fiscal risks, resulting from the absence of explicit allocation in the contract or from opportunities for transferring risk back to government. PFRAM users can review a battery of 11 classes of risk, and information provided by the users is summarized in a project risk matrix. Following a structured questionnaire, PFRAM assists the users in identifying the main risks arising from a PPP project, their allocation, likelihood, impact, as well as potential mitigation measures. Box 16 presents an illustrative fiscal risks matrix for the Jamaica Highway 2000 project.

In addition, resiliency planning is needed to ensure that uncertainties about future climate impacts are incorporated in today’s investment decision in the Caribbean region (Public-Private Infrastructure Advisory Facility 2016). Given the long lifespan of most infrastructure, governments in the region and private sector must work together to incorporate elements of comprehensive resiliency to natural disasters and climate change into the design, construction, financing, and operation and maintenance of all future infrastructure projects—whether procured through traditional procurement or PPPs. A wide range of options are available to increase the resiliency of infrastructure to natural disasters and climate change, including engineering options, technological and ecosystem-based options, and institutional arrangements.

Risks associated with natural disasters and climate change could be shared and innovative insurance mechanisms used to reduce government exposure. At the project identification stage, in addition to assessing the sources of revenue linked with the affordability of the project, the government must undertake a broad assessment of the risk that arises from potential natural disasters and climate change. Failure to do so may have financial implication for the private sector as well as the public sector, which is usually left to manage unaccounted risks. Parametric insurance can also be used, as it allows to take into account country-specific exposure to natural disaster by paying out in the event of a natural disaster that exceeds a prespecified severity (Alleyne and others 2017).

Encouraging the private partner to insure itself against natural disasters and ensuring public assets could also help. Disaster risk insurance and related hedging tools also help protect governments from the economic burden of disasters and increase the capacity to respond (Alleyne and others 2017). Governments can insure public assets and encourage insurance of private assets to reduce uncertainties associated with direct exposure to disaster risks, although small Caribbean countries can be exposed to high costs as their insurance market is underdeveloped (Alleyne and others 2017).

Governments in the Caribbean region should develop national adaptation policy frameworks for managing risks from natural disasters, including for PPPs (Public-Private Infrastructure Advisory Facility 2016). Given the exposure of Caribbean countries to natural disasters, managing risks related to these events is especially important. Some countries in the region already have policy frameworks for PPPs that include clauses related to force majeure, which typically includes “Act of God”–encompassing extreme climate events (such as natural disasters). But these frameworks do not capture the frequency and magnitude of damage associated with impact of climate change—a more permanent feature. In this context, PPP frameworks in the region should include explicit clauses on climate change to take into account exposure to climate change and include explicit references to making projects resilient to changing climatic conditions.

Box 11.

PPP Fiscal Risk Assessment Model Assessment: Barbados Solid Waste Management Facility

In 2009, the Government of Barbados (GoB) signed a 20-year build, own, operate, and transfer agreement for a solid waste management facility. The GoB provided the land, and its private partner invested in site improvements and equipment and runs the site. The GoB agreed to pay a tipping fee for each ton of acceptable waste delivered to the site, which is adjusted in line with the price index, and guaranteed a minimum monthly and annual tonnage of acceptable waste to be delivered.

The PFRAM tool was used to assess the potential fiscal impact from the main operation of the PPP, the management of acceptable waste. The assessment looked into the minimum cost to the government and the risk in case of higher demand. The PFRAM tool requires a set of financial and other project-related information to perform the assessment. The Barbados waste management PPP contract provided only part of the information required for the PFRAM. Additional inputs needed for the assessment are derived from sources other than the contract to make assumptions that allow to fill the data gaps.¹ Therefore, this assessment is illustrative.

The PFRAM assessment showed that the expected impact of the project on overall government finances was relatively small with an increase in the overall debt level of less than 1.5 percent of GDP at any given time of the project lifecycle. However, the contingent liability from the revenue guarantee amounted to about 5 percent of GDP at the beginning of the project (Figure 21).

View Full Size

Impact of Public-Private Partnership Project on Government Finances

(Percent of GDP)

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Source: Barbados authorities; Jamaican authorities; and IMF staff.Note: The PPP Fiscal Risk Assessment Model allows to check the sustainability of the business model. The tool gives an indication of the level of maintenance and operation costs that can be supported by the expected revenues, given the financing costs. By comparing the calculated maintenance and operation costs with the cost expected by the private partner, the validity of the business model can be assessed (Figure 22).

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Cash Flows of the PPP Private Project Company

(Millions, national currency)

Citation: Departmental Papers 2019, 007; 10.5089/9781498307062.087.A001

Sources: Barbados authorities; and IMF staff.Note: As not all risks can be modeled in the PPP Fiscal Risk Assessment Model; certain upside and downside risks entailed in the project have to be assessed outside the model. Upside risks include the production of compost of predefined quality, the sales revenues of which are shared between private company and the government, and revenues from the sale of any carbon credits, which would go to the government. Downside risks, on the other hand, include an unforeseen surge in demand for waste disposal, which would raise the fees that the government must pay to the private partner, and higher than expected amounts of nonacceptable waste, which the government would have to take care of.

• Download Figure
• Download figure as PowerPoint slide

¹These include (1) Inter-American Development Bank study (Riquelme, Méndez, and Smith 2016) on waste management in the region, which has been taken as starting point for estimating the volume of acceptable waste delivered to site, with GDP growth used as driver for demand; (2) the investment by private partner (that is, construction cost and investment in equipment) has been derived from transfer value stated in contract, taking into account depreciation; (3) operation and maintenance cost were set at a level that made the project viable under minimum volume; (4) the financing structure used by the private partner is assumed to be split between debt and equity similarly to the financing mix of the mother company; and (5) financing cost are based on industry and risk assumptions.

Box 16.

Fiscal Risk Matrix of Jamaica Highway 2000—Phase I

An assessment of fiscal risks from the Jamaican Highway 2000 project was guided by the PFRAM risk matrix (Table 4). The evaluation was done after the project was already in operation and the assessment has been impacted by the nonmaterialization of most risks. The PPP contract allocated key risks partly to the public and partly to the private partner. Several risks were also shared between the partners. The review of the allocation of subrisk reveals that the risk allocation was also largely in line with common practice.

All but two risks were assessed as having a low likelihood. Considering the likelihood of the risk materializing and the potential fiscal impact, most risks received a “low” risk rating or were identified as irrelevant. Only the financial risk and the renegotiation risk were rated as “medium,” and as there were no mitigating measures in place for these two risks, it was suggested that these were areas of high priority for action.

The financial risk, which had been identified as critical through the assessment, materialized during the contract implementation and required government intervention. After the commercial closure with the preferred bidder, the private party faced difficulties in obtaining the financing in time for meeting the deadline for financial closure. Consequently, the government assisted with the raising of the initial financing.

Table 4.

Risk Matrix for Highway 2000

Project Risk-Sharing Arrangements

Sources: Jamaican authorities; and IMF staff.

Table 4.

Risk Matrix for Highway 2000

Project Risk-Sharing Arrangements

Identification of Risks	Allocation	Likelihood	Fiscal Impact	Risk Rating	Mitigation Strategy	Priority Action
				Likelihood*Impact		Rating*Mitigation
1 Governance risks	Public	Low	Medium	Low	NO	Medium priority
2 Constructon risks	Shared	Low	Medium	Low	YES	Low priority
3 Demand risks	Private	Low	Low	Irrelevant	YES	NO action required
4 Operatonal and performance risks	Private	Low	Low	Irrelevant	YES	NO action required
5 Financial risks	Shared	Medium	Medium	Medium	NO	High priority
6 Force majeure	Shared	Low	Low	Irrelevant	YES	NO action required
7 Material adverse government actons	Public	Low	Medium	Low	NO	Medium priority
8 Change in law	Public	Low	Medium	Low	NO	NO action required
9 Rebalancing of fnancial equilibrium	Private	Low	Medium	Low	NO	Medium priority
10 Renegotiation	Public	Medium	Medium	Medium	NO	High priority
11 Contract terminaton	Shared	Low	Medium	Low	NO	Medium priority

Sources: Jamaican authorities; and IMF staff.

View Table

Table 4.

Risk Matrix for Highway 2000

Project Risk-Sharing Arrangements

Identification of Risks	Allocation	Likelihood	Fiscal Impact	Risk Rating	Mitigation Strategy	Priority Action
				Likelihood*Impact		Rating*Mitigation
1 Governance risks	Public	Low	Medium	Low	NO	Medium priority
2 Constructon risks	Shared	Low	Medium	Low	YES	Low priority
3 Demand risks	Private	Low	Low	Irrelevant	YES	NO action required
4 Operatonal and performance risks	Private	Low	Low	Irrelevant	YES	NO action required
5 Financial risks	Shared	Medium	Medium	Medium	NO	High priority
6 Force majeure	Shared	Low	Low	Irrelevant	YES	NO action required
7 Material adverse government actons	Public	Low	Medium	Low	NO	Medium priority
8 Change in law	Public	Low	Medium	Low	NO	NO action required
9 Rebalancing of fnancial equilibrium	Private	Low	Medium	Low	NO	Medium priority
10 Renegotiation	Public	Medium	Medium	Medium	NO	High priority
11 Contract terminaton	Shared	Low	Medium	Low	NO	Medium priority

Sources: Jamaican authorities; and IMF staff.

Sources: Jamaican authorities; and IMF staff.