Abstract
The 2012 Article IV Consultation with Liberia discusses the economic developments and policies of the country. Liberia recorded strong macroeconomic performance under the three-year Extended Credit Facility (ECF) Arrangement, but poverty continued to be pervasive. The short- to medium-term outlook has remained favorable, although subject to considerable risks. Following resumption of iron ore exports in 2011, real GDP growth is estimated at 9 percent in 2012, supported by strong growth in the mining sector and expansionary fiscal policy for infrastructure investment. IMF staff supports the authorities’ request for a successor arrangement under the ECF.
Liberia: The Investment-Financing-Growth Nexus1
Introduction
1. The Government of Liberia has set an ambitious agenda to transform the Liberian economy and to reach middle-income country status by 2030, the underlying goal of their second Poverty Reduction Strategy (PRS2). The authorities’ strategy focuses on achieving accelerated broad-based growth by scaling up investment in infrastructure and human capital. As in other low-income countries, the question is how fast public investment should be scaled up to address massive infrastructure gaps and developmental needs and the trade-offs and potential risks associated with different financing options and the required changes in fiscal policy.
2. To address this question, we examine some preliminary results from simulations of an inter-temporal macroeconomic model used to explore the nexus between public investment, financing, and growth. The model is designed to capture the features of low income countries (LICs) like Liberia and, where possible, has been calibrated using Liberian economic data. The analysis presented here complements the standard IMF-World Bank debt sustainability analysis.
3. The model used in this note assumes that public investment has a high economic and social rate of return and is highly complementary toward private sector investment. An increase in public investment therefore provides both a direct and an indirect boost to GDP growth. The model accommodates different methods of financing the investment buildup. The government may increase taxes (or reduce transfer payment), borrow from domestic markets, or borrow on international commercial markets. Concessional financing, which has so far been the primary source for Liberia, is included as an exogenous input to the model. The model also incorporates assumptions about efficiency of public expenditure (i.e. the rate at which executed investment translates into productive capital). For different levels of public investment, and assumptions about minimum available financing and efficiency of public investment, the model generates different financing options, the size of the fiscal adjustment needed, and the trade-offs on the long term path of public debt and growth.
4. The exercise yields three main policy conclusions. First, the efficiency of public investment is important for raising the growth dividend from higher investment spending. Second, the rapid investment strategy (one in which investment is scaled up by an annual average of 14 percent of GDP) will cause debt to rise to over 100 percent of GDP if the government is not able to secure enough grants or to create enough fiscal space. Finally, a moderate to high scaling up of investment (by an annual average of 9 percent of GDP) is feasible, though it may still push debt to near its statutory limit and must be planned carefully.
5. The note is organized as follows: the first section provides some background on Liberia, focusing on the infrastructure and human capital needs and the government’s strategy to address these needs. The second section provides a brief, non-technical overview of the model. The third section discusses the simulations, their key results, and the tradeoffs involved in each. The note ends with some concluding remarks.
I. Background
6. Liberia’s infrastructure suffered major damage during a 17-year civil war. The country’s hydropower plant was virtually destroyed leaving the country in darkness. Roads were neglected, bringing trade to a complete halt while limiting the population’s access to public services. Schools and health facilities were damaged and public institutions were abandoned, greatly affecting the country’s standard of living. Along with a deteriorated infrastructure, Liberia’s managerial capacity to execute infrastructure projects was completely undermined as qualified staff fled the country.
7. Since the signing of the Peace Accords in 2003, Liberia has recorded solid economic growth averaging 7.4 percent over this nine-year period. While this result may well reflect an initial boost associated with the recovery after the massive losses during civil war, since 2009 real economic growth has averaged 7 percent driven mostly by non-mining activities. Going forward, it is expected that the mining sector will become a more important source of growth, which is estimated to stabilize at around 6 percent.
8. While public investment in Liberia reached 4.2 percent of GDP in FY 2011/12, it has been low—around 3 percent of GDP on average in the previous five years—and below that of other low income countries (LICs). Furthermore, Liberia also has a weak track record of project implementation with an execution rate of around 60-65 percent of budgeted capital expenditures reflecting legal, institutional, and managerial bottlenecks which hamper the efficient selection and effective implementation of public investment projects.
Public Investment Across Countries
(Percent of GDP)
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
Source: WEO database and IMF staff calculations.9. Liberia is a fragile state with massive infrastructure gaps and developmental needs:
Per capita income is very low with 84 percent of the population in poverty ($1.25 per day);
Employment creation has been low; according to the 2010 Liberia Labor Force Survey, 68 percent of employed Liberians work in the informal sector;
Child mortality rate, under five, is 103 per 1,000, while the maternal mortality rate is high at 770 per 100,000 births, both of which are high compared to neighboring countries;
Pipe-borne water—the main source of drinking water—is low and water deficiency is about 61 percent of the population. Improved sanitation (percent of population with access) is very low at only 17 percent of the population;
Years of war affected the education system which needs to be rebuilt to deliver quality practical and vocational training. The adult literacy rate (percent of people ages 15+) is 59 percent;
Liberia ranks at the bottom of the UN’s Human Development Index (182 out of 187 countries in 2011);
Infrastructure is weak. The country currently produces 23MW of electricity covering just 2–3 percent of the population and at the highest price in the world, access to electricity is also a constraint to small-scale manufacturing and industrial enterprises; poor roads are an impediment to trade (both domestic and cross-border) and a constraint to access public services such as education and health, especially during the rainy season;
Aid flows are around 39 percent of GDP, including some 2 percent of GDP for budget support.
10. Liberia successfully reached the completion point under the HIPC initiative in June 2010, helping the country to bring down its external debt from close to 150 percent of GDP to less than 9 percent. The government’s tax collections have increased substantially over the last six years (from 11 percent of GDP in FY 2005/06 to 22 percent of GDP in 2011/12) which together with the fiscal space created by the debt relief provides an important opportunity to increase public investment.
11. The government has laid out a five-year development plan in support of the country’s goal to achieve middle income status by 2030 incorporated in its PRS2. The plan focuses on five strategic pillars—at an estimated cost of $3.3 billion over FY 2012/13 to FY 2016/17—to increase productivity, boost economic growth, and improve social inclusion—particularly among youth. The pillars are
Economic transformation through investing in infrastructure, particularly rehabilitation of the hydropower plant and related transmission and distribution networks destroyed during the civil war, improvement of trunk roads, and rehabilitation and expansion of ports, and improvement in the information and communications technology. Total infrastructure investment needed is estimated at $2.2 billion, including $1.4 billion for roads and $0.5 billion in the energy sector, of which $0.2 billion is needed for the hydropower plant.
Human development. The strategy aims to increase access and improve the quality of education and health, provide social protection for vulnerable persons, and rehabilitate and expand the infrastructure in water and sanitation. The estimated cost is $0.5 billion.
Peace, security, and the rule of law. The Government will need to invest in the security sector to maintain a stable and peaceful environment as UN forces (UNMIL) drawdown. The expected cost is $0.4 billion.
Governance and public institutions. The government plans to invest US $0.1 billion on the public sector modernization and reform agenda, decentralization and local governance, efforts and enhancing transparency and accountability.
Cross cutting issues. The government intends to invest some $0.1 billion to develop youth skills and increase empowerment, child protection, gender equality, and human rights’ protection.
Liberia: Agenda for Transformation Costing Summary
(US$s millions)
Liberia: Agenda for Transformation Costing Summary
(US$s millions)
| FY 12/13 | FY 13/14 | FY 14/15 | FY 15/16 | FY 16/17 | Five Year | |
|---|---|---|---|---|---|---|
| Pillar 1. Economic transformation | 594.1 | 532.4 | 439.4 | 354.9 | 267.8 | 2,188.6 |
| Pillar 2. Human development | 87.2 | 100.9 | 120.8 | 121.1 | 128.9 | 558.9 |
| Pillar 3. Peace, security, and rule of law | 73.1 | 90.3 | 92.1 | 77.1 | 73.0 | 405.6 |
| Pillar 4. Governance and public institutions | 40.3 | 16.5 | 14.5 | 14.1 | 10.0 | 95.5 |
| Pillar 5. Cross-cutting issues | 19.5 | 28.7 | 22.5 | 22.1 | 19.2 | 111.9 |
| Total | 814.2 | 768.9 | 689.3 | 589.2 | 498.9 | 3,360.5 |
Liberia: Agenda for Transformation Costing Summary
(US$s millions)
| FY 12/13 | FY 13/14 | FY 14/15 | FY 15/16 | FY 16/17 | Five Year | |
|---|---|---|---|---|---|---|
| Pillar 1. Economic transformation | 594.1 | 532.4 | 439.4 | 354.9 | 267.8 | 2,188.6 |
| Pillar 2. Human development | 87.2 | 100.9 | 120.8 | 121.1 | 128.9 | 558.9 |
| Pillar 3. Peace, security, and rule of law | 73.1 | 90.3 | 92.1 | 77.1 | 73.0 | 405.6 |
| Pillar 4. Governance and public institutions | 40.3 | 16.5 | 14.5 | 14.1 | 10.0 | 95.5 |
| Pillar 5. Cross-cutting issues | 19.5 | 28.7 | 22.5 | 22.1 | 19.2 | 111.9 |
| Total | 814.2 | 768.9 | 689.3 | 589.2 | 498.9 | 3,360.5 |
It is estimated that the government will contribute between 12–15 percent of the financing required to fully implement the investment program. The remaining 85–88 percent is still unfunded. The authorities are planning a pledging donors’ conference in late 2012 to secure both loans and grants to fully execute its development strategy. The model below can be used to assess the trade-offs of alternative financing options for PRS2.
II. The Model
12. The model is a small two-sector open economy framework. A complete technical explanation of the model can be found in Buffie et al. (2012). What follows is an outline of the key features of the model.
Consumers
13. The economy is populated with two types of agents: savers, who have access to capital markets and inter-temporally optimize consumption and labor, and non-savers, who operate by a rule of thumb and simply consume what they earn each period.
Savers
14. Each period, savers maximize the present discounted value of their lifetime utility (which is just a function of consumption) by choosing how much to consume, how much to invest, and how much debt to incur, all subject to a budget constraint.
15. Savers receive income from: renting capital to firms (both traded- and non-traded goods firms); wages earned on the labor market; remittances, which act as transfer payment from the rest of the world; and net government transfers (other taxes net of current spending). They pay for: a constant elasticity of substitution (CES) basket of consumption goods; a tax on consumption; investment in new capital as well as the adjustment cost of changing the capital stock; the cost of servicing debt (both domestic and foreign) and adjusting their debt portfolio; and user fees on public infrastructure.
16. The savers’ optimization problem yields standard results. The path of consumption depends on the real interest rate and the level of the tax on consumption. There are also three arbitrage conditions. The real interest rate on domestic bonds is equal to the real interest rate on foreign bonds, adjusted by portfolio costs. Additionally, the return on capital in the traded and non-traded sector, net of capital adjustment costs, must equal the real interest rate.
17. The portfolio adjustment costs allow the model to capture an important feature of the market for debt in LICs. Steep adjustment costs reflect the immobility of capital in LICs, where the private sector has a limited ability to borrow overseas and the capital account is fairly closed.
Non-Savers
18. Savers are additionally constrained by an inability to access capital markets, and thus they do not smooth consumption over time. Because there is no formal optimization problem for non-savers, their labor supply is fixed in each period. Income comes from wages, remittances, and government transfers. Each period, non-savers purchase a CES basket of goods and consume it in its entirety.
19. We incorporate non-saving households to add an element of reality to the model, given the high proportion of households in LICs that consume hand-to-mouth. The addition of these rule-of-thumb consumers into the model allows for realistic non-Ricardian outcomes.
Firms
20. The economy is divided into traded and non-traded goods sectors, each populated by representative firms. These firms use Cobb-Douglas technology to produce output taking as inputs labor, private capital, and effectively-produced public capital:
Here, q is the quantity of goods produced, A is aggregate factor productivity, z is public investment, k is private capital, and L is labor. Each of these factors is sector-specific (denoted by the subscript j).
21. Inputs to production are bought in competitive markets, so profit-maximizing firms pay the marginal products for each. Labor is inter-sectorally mobile, but capital is sector-specific. In equilibrium, the rental rate for each type of capital is the same, though the prices will not follow the same transition paths.
22. The form of the production function captures the complementarities between public and private capital. All else being equal, as the stock of effective public capital increases, the return on private investment and labor both increase. For any investment scaling-up, then, GDP rises by accounting identity logic, where government investment is an additive component of output, but also by increasing the incentives for private investment and labor to rise as well.
The Government
Public Investment
23. Public investment can be thought of as a two-step process. In the first step, the government invests in public infrastructure according to a standard accumulation equation:
Here, δ is the depreciation rate of public capital, and i is the level of investment in capital. In the second step, a portion of the public infrastructure is converted into effectively productive capital that firms use as an input to production:
The bar denotes variables at initial steady state and s is a parameter that governs the portion of infrastructure that is converted to effectively productive capital. Given that s lies on the unit interval, this feature of the model implies that one dollar of additional public investment translates to less than one dollar of effectively produced capital.2 Note that only effective public capital increases the return on private investment by entering into the production function.
The Budget Constraint
24. The government budget constraint requires that spending on transfers (current spending), debt service, and infrastructure investment not exceed revenue collected from a consumption tax and from user fees on infrastructure.3 When revenues do not meet expenditures, the resulting fiscal gap is financed through domestic borrowing, external concessional borrowing, or external commercial borrowing. External grants can also be used to finance the budget. Below, we show simulations with realistic assumptions about grants to explore their effect on the debt profile required for investment.
25. The expenditure side of the budget constraint also contains a term meant to capture cost overruns on public infrastructure investment. LICs have a shortage of skilled administrators, leading to public projects that are slowed by poor planning, weak oversight, or coordination problems.
26. The interest rate on domestic debt is (quite obviously) the domestic interest rate. The interest rate on concessional debt is assumed to be a constant negotiated rate. The interest rate on external commercial debt varies proportionally to a risk premium, defined as the ratio of the external public debt-to-GDP ratio to its initial steady state.
Fiscal Adjustment
27. The difference between revenues and expenditures is the fiscal gap, which can be financed in the short run by borrowing or by fiscal adjustment (i.e., taxes and transfers). In the long run, though, the tax on consumption and transfer spending must adjust to cover the entire gap.
28. The core policy dilemma facing policymakers is the speed with which this fiscal gap is covered. Sharp increases in the consumption tax or cuts in transfer payments are obviously painful, and policymakers would like to avoid them. If the government does not react quickly enough, or if tax increases or spending cuts are capped at restrictively low levels, interest payments on the debt will rise faster than net government revenue, causing debt to increase explosively.
29. The table below shows the main variables for Liberia. Real per capita economic growth is assumed to be 3 percent, which reflects the average for 2009–11, which was selected to be representative as it includes two consecutive years of slow growth (2009–2010) and one year with rapid growth (2011). Given information for Liberia is limited, the value added in the non-tradable sector was set at 49.4 percent as suggested by the model.
Liberia: Steady-State Macroeconomic Variables
(In percent of GDP, unless otherwise specified)
Average real per capita economic growth over the 2009-11 period equals 2.7 percent and 3.4 percent w hen the 2012 estimate is included.
For debt sustainability analysis, it excludes central government debt to the Central Bank of Liberia.
Liberia: Steady-State Macroeconomic Variables
(In percent of GDP, unless otherwise specified)
| National accounts | |||||
|---|---|---|---|---|---|
| Real economic growth (annual growth rate, in percent) | 8.2 | ||||
| Real per capita growth (annual growth rate, in percent) 1/ | 3.0 | ||||
| Value added in the nontradable sector (in percent) | 49.4 | ||||
| Government accounts | |||||
| Taxes | 21.6 | ||||
| General sales tax rate (in percent) | 7.0 | ||||
| Grants for budget support | 1.7 | ||||
| Spending (net of interest) | 31.0 | ||||
| Interest payments | 0.3 | ||||
| Capital spending | 4.1 | ||||
| Total debt 2/ | 11.7 | ||||
| External sector | |||||
| Current account | −34.1 | ||||
| Trade | −40.7 | ||||
| Exports | 24.6 | ||||
| Imports | −65.4 | ||||
| Remittances | 5.4 | ||||
| External debt | 10.0 | ||||
Average real per capita economic growth over the 2009-11 period equals 2.7 percent and 3.4 percent w hen the 2012 estimate is included.
For debt sustainability analysis, it excludes central government debt to the Central Bank of Liberia.
Liberia: Steady-State Macroeconomic Variables
(In percent of GDP, unless otherwise specified)
| National accounts | |||||
|---|---|---|---|---|---|
| Real economic growth (annual growth rate, in percent) | 8.2 | ||||
| Real per capita growth (annual growth rate, in percent) 1/ | 3.0 | ||||
| Value added in the nontradable sector (in percent) | 49.4 | ||||
| Government accounts | |||||
| Taxes | 21.6 | ||||
| General sales tax rate (in percent) | 7.0 | ||||
| Grants for budget support | 1.7 | ||||
| Spending (net of interest) | 31.0 | ||||
| Interest payments | 0.3 | ||||
| Capital spending | 4.1 | ||||
| Total debt 2/ | 11.7 | ||||
| External sector | |||||
| Current account | −34.1 | ||||
| Trade | −40.7 | ||||
| Exports | 24.6 | ||||
| Imports | −65.4 | ||||
| Remittances | 5.4 | ||||
| External debt | 10.0 | ||||
Average real per capita economic growth over the 2009-11 period equals 2.7 percent and 3.4 percent w hen the 2012 estimate is included.
For debt sustainability analysis, it excludes central government debt to the Central Bank of Liberia.
III. Model Simulation and Results
30. A set of twelve simulations (see attached charts) demonstrate the impact on growth and debt levels of different assumptions regarding the investment path and its financing. Starting from the steady state level of public investment of 4 percent of GDP, the baseline assumes a moderate scaling up of public investment (by 5.5 percentage points of GDP per year over seven years). We also tested for intermediate scaling up of public investment (by 9 percentage points of GDP per year) and a rapid scaling up (by an annual average of 14 percentage points of GDP per year). In turn, each of the three scaling up assumptions were tested for different financing options and different assumed efficiency of public investment. The twelve scenarios are as follows:
Baseline (moderate) scaling-up in which the public investment efficiency rate is 60 percent: the financing needs are closed with concessional borrowing.
Baseline (moderate) scaling-up in which the public investment efficiency rate is 60 percent: the financing needs are closed with concessional borrowing and some domestic financing.
Baseline (moderate) scaling-up in which the public investment efficiency rate increases to 80 percent: the fiscal gap is still closed with concessional borrowing.
Baseline (moderate) scaling-up in which the public investment efficiency rate is only 20 percent: the fiscal gap is still closed with concessional borrowing.
Intermediate scaling-up: fiscal gap is closed with concessional borrowing.
Intermediate scaling-up: fiscal gap is closed with a mix of concessional borrowing and an endogenous tax adjustment.
Intermediate scaling-up: fiscal gap is closed with a mix of concessional borrowing and an endogenous transfer adjustment.
Rapid scaling-up: fiscal gap is closed with additional concessional borrowing.
Rapid scaling-up: fiscal gap is closed with commercial borrowing.
Rapid scaling-up: fiscal gap is closed with a mix of concessional borrowing and commercial borrowing.
Rapid scaling-up: fiscal gap is closed with additional grant aid and a tax adjustment.
Rapid scaling-up: fiscal gap is closed with additional grant aid, commercial borrowing, and a tax adjustment.
31. The table below shows the cumulative boost to real per capita GDP growth from each of the scenarios relative to the assumed steady-state growth rate of 3 percent, as well as the impact of the scenarios on public debt. For assessment purposes, we present here only those scenarios which are most interesting for policy consideration and focus the analysis on the investment-financing-growth nexus.
Liberia: Growth Dividend and Impact on Public Debt from Alternative Investment Levels and Financing Strategies
(Cummulative boost to real per capita growth, in percentage points over a 10 year horizon; and percent of GDP)
Liberia: Growth Dividend and Impact on Public Debt from Alternative Investment Levels and Financing Strategies
(Cummulative boost to real per capita growth, in percentage points over a 10 year horizon; and percent of GDP)
| Investment scenarios and financing strategies | Growth dividend | Public debt (percent of GDP) |
|---|---|---|
| a. Moderate (baseline) scaling-up: fiscal gap closed with additional concessional borrowing | 9.0 | 42.2 |
| b. Moderate (baseline) scaling-up: fiscal gap closed with concessional and domestic borrowing | 6.7 | 45.7 |
| c. Moderate (baseline) scaling-up: fiscal gap closed with additional concessional borrowing (high efficiency) | 13.2 | 41.3 |
| d. Moderate (baseline) scaling-up: fiscal gap closed with additional concessional borrowing (low efficiency) | 3.5 | 42.8 |
| e. Intermediate scaling-up: fiscal gap closed with additional concessional borrowing | 13.1 | 49.4 |
| f. Intermediate scaling-up: fiscal gap closed with a mix of concessional borrowing and an endogenous tax adjustment | 10.5 | 27.1 |
| g. Intermediate scaling-up: fiscal gap closed with a mix of concessional borrowing and an endogenous transfer adjustment | 8.6 | 27.3 |
| h. Rapid scaling-up: fiscal gap closed with additional concessional borrowing | 19.3 | 97.4 |
| i. Rapid scaling-up: fiscal gap closed with commercial borrowing | 20.6 | 108.4 |
| j. Rapid scaling-up: fiscal gap closed with a mix of concessional borrowing and commercial borrowing | 20.1 | 104.8 |
| k. Rapid scaling-up: fiscal gap closed with additional grant aid and a tax adjustment | 15.6 | 51.8 |
| l. Rapid scaling-up: fiscal gap closed with additional grant aid, commercial borrowing, and a tax adjustment | 18.3 | 72.7 |
Liberia: Growth Dividend and Impact on Public Debt from Alternative Investment Levels and Financing Strategies
(Cummulative boost to real per capita growth, in percentage points over a 10 year horizon; and percent of GDP)
| Investment scenarios and financing strategies | Growth dividend | Public debt (percent of GDP) |
|---|---|---|
| a. Moderate (baseline) scaling-up: fiscal gap closed with additional concessional borrowing | 9.0 | 42.2 |
| b. Moderate (baseline) scaling-up: fiscal gap closed with concessional and domestic borrowing | 6.7 | 45.7 |
| c. Moderate (baseline) scaling-up: fiscal gap closed with additional concessional borrowing (high efficiency) | 13.2 | 41.3 |
| d. Moderate (baseline) scaling-up: fiscal gap closed with additional concessional borrowing (low efficiency) | 3.5 | 42.8 |
| e. Intermediate scaling-up: fiscal gap closed with additional concessional borrowing | 13.1 | 49.4 |
| f. Intermediate scaling-up: fiscal gap closed with a mix of concessional borrowing and an endogenous tax adjustment | 10.5 | 27.1 |
| g. Intermediate scaling-up: fiscal gap closed with a mix of concessional borrowing and an endogenous transfer adjustment | 8.6 | 27.3 |
| h. Rapid scaling-up: fiscal gap closed with additional concessional borrowing | 19.3 | 97.4 |
| i. Rapid scaling-up: fiscal gap closed with commercial borrowing | 20.6 | 108.4 |
| j. Rapid scaling-up: fiscal gap closed with a mix of concessional borrowing and commercial borrowing | 20.1 | 104.8 |
| k. Rapid scaling-up: fiscal gap closed with additional grant aid and a tax adjustment | 15.6 | 51.8 |
| l. Rapid scaling-up: fiscal gap closed with additional grant aid, commercial borrowing, and a tax adjustment | 18.3 | 72.7 |
32. The baseline scenario (moderate scaling up of public investment) assumes that concessional borrowing is the main source of financing and there are no changes in tax policies or changes in the current level of grants for budget support. Under this scenario, an average increase of 5.5 percentage points of GDP in investment per annum over a seven year period contributes to an approximately 1 percentage point increase in the annual growth rate of real per capita income on average over ten years. The growth effect peaks 3–4 years after the initial investment and then gradually declines over time returning to trend growth. The cumulative growth dividend to year-on-year per capita growth equals 9 percent over a 10-year horizon with central government total debt estimated to reach approximately 42 percent of GDP at the end of year 10 compared to 11.7 percent of GDP in FY2011/12.
33. We tested for the impact on growth and the debt dynamics of including domestic financing as a source of financing public investment (scenario b). In line with government policies in Liberia, we assumed that the government is able to raise, on an annual basis, 1 percent of GDP in funds in the domestic market. Under this scenario, the crowding out of private investment limits the cumulative growth dividend over a 10 year period to 6.7 percent and worsens the country’s debt dynamics.
Liberia: Public Investment and Real Per Capita Income Growth Under a Moderate Scaling Up Strategy and Different Financing Strategies
(Percentage change, y.o.y.)
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
Sources: IMF staff calculations.Liberia: Public Sector Debt Dynamics Under a Moderate Scaling Up Scenario and Different Financing Strategies
(Percentage of GDP)
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
Sources: IMF staff calculations.34. We also ran scenarios on the baseline for changes in the efficiency of public investment. Several studies show that the quality of projects is critical to translate public investment into productive capital to boost growth. Calderon and Serven (2009) found the quality of public investment contributed positively to per capita growth in advanced and Middle-East and North African countries while Sub-Saharan African countries show poor performance as many projects were not properly assessed or executed.
Changes in growth per capita due to infrastructure investment
(In percent)
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
Sources: Calderon-Serven, 2009.35. From an assumed efficiency rate of public investment of 60 percent in the baseline scenario—the average for low income countries—we tested for higher efficiency rate of 80 percent (scenario c). If the government can increase the efficiency of its public investment, this can further boost economic growth and moderately improve debt dynamics. Under the higher efficiency scenario, real per capita income growth could potentially increase by an additional half percentage point over the medium term. Getting to higher growth faster, in turn, would improve Liberia’s debt dynamics. On the other hand, the same amount of investment with a lower efficiency rate of 20 percent (scenario d) would result in lower per capita growth and moderately weaker debt dynamics. These results point to the critical need for stepped up efforts to tackle bottlenecks to project implementation in Liberia.
Liberia: Real Per Capita Income Growth
(Percentage change, y.o.y)
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
Sources: IMF staff calculations.Liberia: Public Debt Under Different Growth Scenarios
(Percentage of GDP)
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
Sources: IMF staff calculations.36. The results of other scenarios are as follows. In scenario e (intermediate investment strategy), the government is able to secure new grants for budget support equivalent to an additional annual average of 2 percent of GDP to help finance the medium term investment plan. Still, the government needs to continue relying on concessional borrowing within the limits set by the Public Financial Management Act Regulations, which sets the ceiling on public debt to 60 percent of GDP. Under this scenario, the growth dividend to per capita GDP over a 10 year period is 13.1 percent, very similar to the high efficiency rate simulation.
37. Scenario f (intermediate investment strategy) assumes the financing gap is closed by a combination of additional grants and changes in taxes. Under this strategy, the cumulative boost to real per capita growth is approximately 10.5 percent, relatively close to the baseline scenario a. In this case, the population has access to a larger infrastructure level, but the increase in taxes required to close the financing gap constrains consumption, thus limiting the impact on growth.
38. Scenarios h and i represent the government’s rapid investment build-up strategy with the financing gap closed entirely with either concessional or commercial borrowing, respectively. Under these two scenarios, the increase in public investment contributes to a boost in consumption and private investment to the highest possible levels in the medium term, increasing the growth dividend between 19–21 percent. However, this strategy comes at the expense of mounting total public debt to unsustainable levels (between 97–108 percent of GDP), especially in the case of commercial borrowing, which, in the long run, will require large changes in taxes and transfers to service such debt.
39. Scenario k (rapid investment strategy) assumes the government is able to secure 2 percent of GDP in new grants for budget support, but not enough to close the estimated financing gap. In this case, the government is required to adjust the consumption tax from the start while still resorting to some additional concessional borrowing beyond what is assumed in the baseline scaling up scenario. The growth dividend in real per capita GDP over a 10-year period is approximately 15.6 percent with public debt peaking at 52 percent of GDP.
40. The various scenarios have some caveats. For Liberia, several of the parameters were borrowed from previous work on other LICs and Liberia-specific parameters should be treated with caution due to limited data availability. In addition, the model abstracts away from reality (e.g., only two types of consumers, agents with perfect foresight, etc.). Hence, the estimates should be seen only as a first approximation as the model should not be seen as a forecasting tool. Finally, other shocks not discussed here have the potential to change the estimates, either positively or negatively depending on the nature of the shock (i.e., a global recession or a productivity shock in Liberia).
41. Previous work on the investment-growth nexus in Liberia points to a positive but smaller impact from scaling up investment on growth. Yoon and Lane (2009), who focused on the macroeconomic impact of scaling up aid, found that increasing aid would have a positive, albeit limited, effect on the Liberian economy rate of growth. They estimated that increasing foreign aid by 5 percentage points of GDP only contributed to an average increase of 0.16 percent per year in growth over a 10 year period. One factor that contributed to such a lower impact compared to current work is that their analysis assumed that only 30 percent of aid was allocated to investment while the rest was directed to cover current spending, in particular for services paid abroad.
IV. Concluding Remarks
42. The simulations presented here capture some of the costs and benefits of the three potential investment strategies. The model results provide background to support the government’s objectives and expectations to scale up public investment to stimulate the economy while flagging the risks related to debt sustainability.
43. Some caution is required when interpreting the simulations presented here. Liberian data is limited, the model relies on some average parameters calculated for other LICs, and, it abstracts away from reality (e.g., only two types of consumers, agents with perfect foresight, etc.). Thus the results should be interpreted with this in mind. All that being said, the model helps to gauge the relative costs and benefits of various investment and financing scenarios.
44. As expected, the more ambitious investment program yields the larger growth dividend over the medium term, but it comes at the risk of unsustainable debt; hence assessing other options is warranted. The intermediate scaling up option in which the government is able to secure some 2 percentage points of GDP in new grants for budget support results in stronger growth dividends and more stable debt dynamics.
45. In addition to securing the ideal balance of financing options, there is a critical need to increase the efficiency of public investment in Liberia. The government needs to step up efforts to tackle bottlenecks to project implementation and even more importantly to properly prioritize and select high return projects that effectively translate into productive capital.
46. Adjustments in taxes or in current spending (net transfers) may be painful policies which may have high political costs but may be required to successfully implement the government’s PRS2. The authorities are rightly focusing on creating fiscal space to scale up public investment, but further efforts to streamlining current spending are warranted.
References
Briceno-Garmendia, C., K. Smits, and V. Foster, 2008, “Financing Public Infrastructure in Sub-Saharan Africa: Patterns and Emerging Issues,” AICD Background Paper 15 (Washington D.C.: World Bank).
Buffie, E., A. Berg, C. Patillo, R. Portillo, and L. F. Zanna, 2012, “Public Investment, Growth and Debt Sustainability: Putting Together the Pieces,” IMF Working Paper, WP12/144 (Washington, D.C.: International Monetary Fund).
César Calderón, 2009, “Infrastructure and Growth in Africa,” WB Policy Research 4914.
César Calderón and L. Servén, 2009, “Infrastructure and Economic Development in Sub-Saharan.” WB Policy Research, (Washington D.C.: World Bank).
César Calderón and L. Servén, 2010, “Infrastructure in Latin America,” WB Policy Research/5317, (Washington: World Bank).
Dabla-Norris, E., J. Brumby, C. Papageorgiou, A. Kyobe, and Z. Mills, 2011, “Investing in Public Investment: An Index of Public Investment Management Quality,” IMF Working Paper, WP/11/37 (Washington: International Monetary Fund).
Chris Lane and S. Yoon, 2009, “The Macroeconomics of Scaling-up Aid: The Case of Liberia,” IMF (Washington: International Monetary Fund).
Pritchett, L. 2000, “The Tyranny of Concepts: CUDIE (Cumulated, Depreciated, Investment Effort) is not Capital,” Journal of Economic Growth, 5, 361–384.
1. Moderate (Baseline) Scenario
Fiscal Gap Closed w / Additional Concessional Lending
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
2. Moderate (Baseline) Scenario
Fiscal Gap Closed w / Additional Concessional Lending and 1% Domestic Borrowing
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
3. Moderate (Baseline) Scenario – High Investment Efficiency
Fiscal Gap Closed w / Additional Concessional Lending
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
4. Moderate (Baseline) Scenario – Low Investment Efficiency
Fiscal Gap Closed w / Additional Concessional Lending
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
5. Intermediate Investment Scenario
Fiscal Gap Closed w / Additional Concessional Lending
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
6. Intermediate Investment Scenario
Fiscal Gap Closed w / Tax Adjustment
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
7. Intermediate Investment Scenario w / no Grants
Fiscal Gap Closed w / Transfer Adjustment
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
8. Rapid Scaling-Up Scenario
Fiscal Gap Closed w / Additional Concessional Lending
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
9. Rapid Scaling-Up Scenario
Fiscal Gap Closed w / Concessional Lending
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
10. Rapid Scaling-Up Scenario
Fiscal Gap Closed w / Commercial and Concessional Lending
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
11. Rapid Scaling-Up Scenario
Fiscal Gap Closed w / Tax Adjustment
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
12. Rapid Scaling-Up Scenario
Fiscal Gap Closed w / Commercial Borrowing and Tax Adjustment
Citation: IMF Staff Country Reports 2012, 340; 10.5089/9781475542769.002.A004
This note was prepared by Manuel Rosales (AFR) and Will Clark (RES).
Empirical analysis shows that the productivity of infrastructure is high while the return on public investment is low, because not all infrastructure spending becomes public capital. See Hulten (1996) and Pritchett (2000) for details.
User fees capture a fraction of the recurrent costs of maintaining public capital (i.e., maintenance and upkeep caused by capital depreciation). This feature is not meant to represent a public-private partnership model where user fees capture the cost of financing the infrastructure.
