## 3.1 Public Investment and Saving ## Source details **Canonical URL:** [3.1 Public Investment and Saving](https://www.imf.org/-/media/files/publications/wp/2017/wp17233.pdf) ## Other formats - [Markdown version](/-/media/files/publications/wp/2017/wp17233.pdf.md) - [Structured JSON version](/-/media/files/publications/wp/2017/wp17233.pdf.json) --- ### Public investment trends in LIDCs - Analysis based on 47 LIDCs for which the IMF’s World Economic Outlook database contains information on public investment and public saving. - Median public investment in LIDCs: - 5.5 percent of GDP in 2000 - peaked at 7.1 percent of GDP in 2010 - 6.7 percent of GDP in 2015 - declined to 6.4 percent in 2016 - Public investment in LIDCs is higher as a percent of GDP than in emerging and advanced economies and has followed a general upward trend since 2000, first surging before the Global Financial Crisis (GFC) and then picking up again until 2015. - Commodity exporters expanded public investment more than other countries in the pre-crisis period, benefiting from a large terms-of-trade improvement; trends diverged after the GFC with public investment falling in commodity exporters and rising modestly for diversified exporters. - Large cross-country variability: - Examples of substantial scaling-up: Djibouti, Congo, Ethiopia. - Commodity exporters with steady rises until 2014-15: Bolivia, Mongolia, Mozambique, Niger, Tajikistan (followed by a drop in 2014-15). - Significant declines in public investment in Eritrea and Yemen (fragility) and in Nigeria and Uzbekistan (fiscal pressures). - Countries with persistently high levels since 2000: Bhutan (averaged 13 percent of GDP), Vietnam (averaged 9 percent of GDP). - Countries with persistently low levels: Nepal (never exceeding 5 percent of GDP). ### Public saving, fiscal balances, and debt implications - Public saving has generally not been scaled up commensurately with the increase in public investment. - Correlation between changes in public investment and public saving is evident, but investment increases exceeded saving increases in most countries, especially in recent years. - Among the 33 countries where the public investment/GDP ratio increased between 2001-05 and 2011-15, public saving rose in 27, but only in 10 of them did it rise enough to cover the increase in public investment. - Median public saving as a share of GDP: - rose 2.9 percentage points between 2000 and 2007—twice as much as public investment during that period - declined sharply during the GFC, rebounded briefly, then slipped again reflecting lower commodity prices - dropped 2.4 percentage points of GDP since its 2007 peak, returning to early 2000s levels - In 2015: - public investment exceeded public saving in 42 out of 46 LIDCs - the gap between median public investment and median public saving reached 4.8 percent—the widest since 2000 - Resulting fiscal consequences: - negative public saving-investment balances contributed to higher government debt-to-GDP ratios after a notable drop in the 2000s - median general government debt ratio rose from 34 percent in 2013 to 43 percent in 2016 - since 2010 LIDCs issued more than USD 22 billion in sovereign bonds - examples: Ethiopia issued a USD one billion Eurobond in 2014; Senegal issued a USD 1.1 billion Eurobond in May 2017 ### Public infrastructure investment (survey-based) - IMF LIDC country desk survey coverage: - 32 teams provided information on public investment in economic infrastructure over the last five years - 23 had data by sector - For the median LIDC in the sample: - investment in economic infrastructure accounted for about half of total public investment (correlation between public infrastructure investment and total public investment is 0.8 in this sample) - median infrastructure investment level: - around 3 percent of GDP in 2011–14 - dropped below 2½ percent in 2015 as commodity exporters were hit by falling export prices - Cross-group patterns: - frontier market economies had somewhat higher levels of investment - fragile states typically had lower investment levels - Sectoral composition of investment in economic infrastructure (median shares reported): - transportation: about half of total investment in economic infrastructure (usually below 50 percent) - water and sanition: 22 percent - energy: 19 percent - ICT: 6 percent - Observations: - relatively low share of energy investment is troubling given access to electricity as a key constraint to development - private provision of ICT services has allowed governments to spend relatively little in that area ### Private participation in infrastructure (PPPs) - Since 2000 LIDCs accounted for: - 6.5 percent of the value of PPP projects in all emerging market and developing economies - 10.5 percent of the number of PPP projects - PPP flows in LIDCs: - averaged about 0.4 percent of LIDC GDP in the last five years (similar to EMs) - after a sharp acceleration in the early 2010s, PPP flows declined in the most recent years - of the $43 billion in LIDC PPP projects since 2010, more than half has been invested in Asia and one third in Sub-Saharan Africa - Distribution and notable facts: - Vietnam and Bangladesh have the largest number of projects; Lao PDR leads in volume - PPPs have been used for regional projects (example: Central Corridor—integrated transport program across five countries with an investment of about $18 billion) - Project characteristics: - 87 percent greenfield projects since 2000 - 8 percent brownfields - 97 percent of contracts with the central government - variation in project size with very large examples: - coal plant in Laos: $3.7 billion - expansion of the Onne port complex in Nigeria: $2.9 billion - thermal power generation project in Vietnam: $2 billion - nine projects started since 2010 are valued over $1 billion - Role of MDBs: - more than a quarter of projects in LIDCs involve MDB support (direct loans, syndication, equity investment, partial credit guarantees, political risk coverage) - MDB presence is associated with a lower probability that a project comes under distress or is canceled ### Financing for infrastructure: Official Development Finance (ODF) and cross-border lending - Detailed OECD data show: - LIDCs received nearly $17 billion in project finance from MDBs and OECD members in 2014 - 87 percent of ODF for LIDCs consisted of grants and concessional loans (contrast: 56 percent for all developing countries) - bulk of the money went to public projects; direct support to the private sector amounted to $0.9 billion in 2014 - Sectoral shifts in ODF: - share of projects in water and transportation declined steadily since 2006 - share of energy increased to about 30 percent in 2014 - Dispersion across countries: - for all LIDCs in 2014, the median ratio of ODF to GDP equaled 1.3 percent ### Financing patterns and donor composition - Multilateral support accounted for 57 percent of ODF, bilateral for 43 percent. - ODF commitments amounted to around $24 billion in 2014, exceeding disbursements by a wide margin. - According to OECD (2016), ODF covers 6-7 percent of infrastructure investment in developing countries. - Grants accounted for the bulk of financing in fragile states; frontier markets and commodity exporters received less ODF (relative to their GDP) than other country groups due to a higher domestic revenue base and greater access to commercial borrowing. - In the IMF survey, only 40 percent of LIDC country teams indicated that new projects included a budget for maintenance. ### Role of emerging and non-traditional donors - Between 2000 and 2013 almost 60 percent of Chinese-funded projects were infrastructure ones (AidData). - Gutman et al. (2015) calculate that China contributes about 20 percent of external finance for infrastructure projects in Sub-Saharan Africa, with most of that financing provided by China's EXIM Bank. - India’s development financing for infrastructure is estimated at $1.3 billion in 2014, with most of it going to neighboring countries, primarily for energy and transportation. - New multilateral institutions expanding the donor base include the Asian Infrastructure Investment Bank (AIIB) and the New Development Bank (NDB). ### Cross-border bank lending and commercial flows - MDBs participate in about one fourth of international syndicated cross-border loans to LIDCs. - Vietnam, Uzbekistan, Nigeria, Lao PDR, Ethiopia and Kenya are the largest recipients of international syndicated loans. - Total cross-border bank lending rose steadily in the late 2000s, peaking in 2012—when it amounted to about USD 40 billion—before falling significantly alongside the drop in commodity prices in 2014–15. - Since 2007, almost 30 percent of cross-border bank lending in LIDC financed infrastructure projects; the share in EMs is about 22 percent. - In terms of sector distribution, 52 percent of infrastructure loans go to energy and utilities, 19 percent to telecommunications, 17 percent to transportation. - Cross-border bank lending generally represents a complementary source of external financing with respect to ODF. ### Challenges, absorptive capacity, and efficiency - UNCTAD (2014) estimates that attaining the SDGs would require increasing spending on economic infrastructure by USD 0.8 to 1.7 trillion a year from current levels (covering all developing countries). - Over the last two years, public debt levels have risen, external financing conditions have tightened, and growth prospects have weakened for the LIDCs. - Dabla-Norris et al. (2012) show that low-income countries have relatively weak public investment management institutions; improving those institutions could increase considerably the efficiency (the “value for money”) of public investment. - Isham and Kaufmann (1999) show that once the ratio between public investment over GDP is too high (above 10 percent), the increase in public investment is associated with a declining productivity of investment projects. - Presbitero (2016) shows that infrastructure projects undertaken in periods when public investment accelerates compared to its historical patterns are less likely to be successful, indicating absorptive capacity constraints. - The IMF team survey found multiple obstacles to scaling up public investment in economic infrastructure; no single constraint emerged as dominant in the full sample. Subgroups: fragile state desks emphasized availability of external finance and administrative capacity; frontier economies emphasized availability of domestic resources and limits on debt accumulation. ### Private participation, PPPs, and institutional conditions - Even under optimistic assumptions about future improvements in public investment efficiency, domestic resource mobilization, and concessional financing, addressing the infrastructure shortfall requires a significant increase in private sector participation. - In the near future private participation is likely to occur primarily through PPPs, concentrated in the energy sector. - PPP use is correlated with domestic institutions (rule of law, levels of corruption) and with the Infrascope index; the average Infrascope index for LIDCs is significantly lower than for EMs, with particularly large gaps in the legal regulatory framework and the presence of financial facilities. - Macro-fiscal implications of PPP projects could be large and expose countries to fiscal risks; a strong regulatory environment and a robust institutional framework are essential to implement PPPs sustainably. - There is scope to improve collaboration between local governments and MDBs in the preparation, structuring and financing of infrastructure projects (examples: World Bank Global Infrastructure Facility and the EBRD Equity Participation Fund). ### Policy recommendations and way forward - Mobilize domestic resources for public investment by increasing tax revenue and streamlining and prioritizing expenditures. - Increase access to concessional external financing. - Develop local capital markets. - Strengthen the institutional and regulatory framework to expand private sector involvement in the provision and financing of infrastructure investment, supported by multilateral development banks and development finance institutions. - Improve “value for money” in public and PPP investment projects. - Seek financing on the most concessional terms possible where fiscal space exists, with support from the international community. - Scale up infrastructure investment gradually while building capacity and strengthening institutions to avoid surpassing absorptive capacity limits. ### Conclusions — summary of empirical and policy messages - Public investment, including in infrastructure, has broadly increased in LIDCs over the last 15 years, but the quantity, quality and accessibility of infrastructure remain considerably lower than in emerging market economies. - Outside the telecom sector, infrastructure services in LIDCs are primarily provided by the public sector; private participation is largely channeled through PPPs and has declined recently after an early-2000s spike. - Grants and concessional loans from development partners are essential sources of infrastructure funding in LIDCs; international syndicated loans play an important complementary role in a few countries. - Improving LIDC infrastructure to levels consistent with attaining the Sustainable Development Goals requires coordinated action across public investment efficiency, domestic resource mobilization, concessional financing, institutional strengthening, and a substantial increase in private-sector involvement. ### Appendix I — Public Investment Scaling-up in Ethiopia (selected facts) - Public investment rose from 12 percent of GDP in 2009 to 22 percent in 2015. - Infrastructure outcomes (2010–2015): - Power generating capacity more than doubled. - Number of telecom users quadrupled. - Stock of asphalt roads rose by 30 percent. - Number of electric outages doubled between 2011 and 2015. - Reliance of manufacturing firms on own electricity generators doubled between 2011 and 2015. - Financing for capital spending: - External borrowing averaged 5.7 percent of GDP per year over the period 2010-15. - Private banks are required to buy government bonds equivalent to 27 percent of their annual loans to fund long term investments. - Macroeconomic and debt considerations: - Real GDP increased at an average rate of 10 percent per year between 2010 and 2015. - Both domestic and external public debt stood close to 30 percent of GDP in 2015. - The 2015 debt sustainability analysis elevated the risk of debt distress from low to medium. - External creditor composition note: - China accounted for 29 percent of total external borrowing during 2012-2015. ### Appendix II — Hydropower PPPs in Lao PDR (selected facts) - NT2 is a $1.45 billion hydroelectric project that began operation in 2010. - Concession agreement signed in 2002 for 25 years; project to be transferred to the government after concession period. - NTPC shareholders: Electricite de France (EDF) – 35 percent; Lao State Holding Enterprise (25 percent); Electricity Generating Public Company Ltd. (Thai company – 25 percent); Italian-Thai Development Public Company Limited (Thai entity - 15 percent). - Power Purchasing Agreement: Electricity Generating Authority of Thailand (EGAT) agreed to acquire 95 percent of power produced by NT2 for the first 13 years. - Financing: Equity financing amounted to $450 million; remainder financed by debt from a broad base of lenders including two bilateral lenders, five multilateral lenders, four export credit agencies, and 10 commercial banks. - To date NT2 has provided close to $1 billion of export revenue to Laos and close to $180 million in royalties and dividends to the government. ### Appendix III — Solar Micro-Grids in Kenya (selected facts) - Steama.co platform: - Total installed capacity managed under the Steama.co platform is around 200kW. - Serving 1000 households and businesses with close to 10,000 total end-beneficiaries. - Subscription increased from around 100 connections in 2014 to 1000 in 2015; projected to reach 5000 by end-2016. - Number of service providers increased from 3 in 2014 to 6 in 2016. - Number of solar sites tripled to 31 by 2016. - Generation capacity increased by a factor of 50 between 2014 to 2016. - Comparative costs and affordability: - Price per kWh from the solar micro-grid is around US$1.5. - National grid tariff about US$ 0.15 per kWh. - Levelized cost of electricity (LCOE) in Kenya from individual diesel generators is about US$ 2 per kWh. - Kerosene lanterns cost between US$ 5 to 10 per kWh. - Home solar systems cost between US$ 2.5 to 8 per kWh. - Connection fee to micro-grids is about US$10. - Often subsidized national grid connection fees are US$ 150 to $350; the real cost of a single-phase connection is about US$1000. - Typical micro-grid customer annual spending for lighting, powering television and phone charging is $120. - Average annual lighting spending per household in Kenya is $157. - Market potential and investment trends: - Bloomberg New Energy Finance and Lighting Global (2016) forecast: local solar power will reach one in three off-grid households by 2020. - Off-grid solar (micro-grid and home solar systems) will improve access to electricity for 89 million people in sub-Saharan Africa and Asia by 2020. - Investment in off-grid solar in sub-Saharan Africa increased by 15 fold from $18.4 million in 2012 to $276 million in 2015. - Off-grid population in sub-Saharan Africa and Asia spent over $20.6 billion on lighting in 2014, which is between $45 to $186 per household. - Enabling factors and policy/institutional points: - Synergy between technical advances in electricity generation and mobile telephony (GSM coverage enabling SMS prepayment). - Mobile banking and cloud-based metering/billing reduce administrative burdens and cash-collection risks. - Removal of VAT and tariffs for solar imports has reduced the cost of capital. - Micro-grids are cheaper than off-grid alternatives (diesel, kerosene, home solar systems) and charge much lower connection fees than the national grid. *Source: IMF Working Paper — excerpt "3.1 Public Investment and Saving" (wp17233).* ### 3.1 Public Investment and Saving ### 3.1 Public Investment and Saving ### Public investment trends in LIDCs - Analysis based on 47 LIDCs for which the IMF’s World Economic Outlook database contains information on public investment and public saving. - Median public investment in LIDCs: - 5.5 percent of GDP in 2000 - peaked at 7.1 percent of GDP in 2010 - 6.7 percent of GDP in 2015 - declined to 6.4 percent in 2016 - Public investment in LIDCs is higher as a percent of GDP than in emerging and advanced economies and has followed a general upward trend since 2000, first surging before the Global Financial Crisis (GFC) and then picking up again until 2015. - Commodity exporters expanded public investment more than other countries in the pre-crisis period, benefiting from a large terms-of-trade improvement; trends diverged after the GFC with public investment falling in commodity exporters and rising modestly for diversified exporters. - Large cross-country variability: - Examples of substantial scaling-up: Djibouti, Congo, Ethiopia. - Commodity exporters with steady rises until 2014-15: Bolivia, Mongolia, Mozambique, Niger, Tajikistan (followed by a drop in 2014-15). - Significant declines in public investment in Eritrea and Yemen (fragility) and in Nigeria and Uzbekistan (fiscal pressures). - Countries with persistently high levels since 2000: Bhutan (averaged 13 percent of GDP), Vietnam (averaged 9 percent of GDP). - Countries with persistently low levels: Nepal (never exceeding 5 percent of GDP). ### Public saving, fiscal balances, and debt implications - Public saving has generally not been scaled up commensurately with the increase in public investment. - Correlation between changes in public investment and public saving is evident, but investment increases exceeded saving increases in most countries, especially in recent years. - Among the 33 countries where the public investment/GDP ratio increased between 2001-05 and 2011-15, public saving rose in 27, but only in 10 of them did it rise enough to cover the increase in public investment. - Median public saving as a share of GDP: - rose 2.9 percentage points between 2000 and 2007—twice as much as public investment during that period - declined sharply during the GFC, rebounded briefly, then slipped again reflecting lower commodity prices - dropped 2.4 percentage points of GDP since its 2007 peak, returning to early 2000s levels - In 2015: - public investment exceeded public saving in 42 out of 46 LIDCs - the gap between median public investment and median public saving reached 4.8 percent—the widest since 2000 - Resulting fiscal consequences: - negative public saving-investment balances contributed to higher government debt-to-GDP ratios after a notable drop in the 2000s - median general government debt ratio rose from 34 percent in 2013 to 43 percent in 2016 - since 2010 LIDCs issued more than USD 22 billion in sovereign bonds - examples: Ethiopia issued a USD one billion Eurobond in 2014; Senegal issued a USD 1.1 billion Eurobond in May 2017 ### Public infrastructure investment (survey-based) - IMF LIDC country desk survey coverage: - 32 teams provided information on public investment in economic infrastructure over the last five years - 23 had data by sector - For the median LIDC in the sample: - investment in economic infrastructure accounted for about half of total public investment (correlation between public infrastructure investment and total public investment is 0.8 in this sample) - median infrastructure investment level: - around 3 percent of GDP in 2011–14 - dropped below 2½ percent in 2015 as commodity exporters were hit by falling export prices - Cross-group patterns: - frontier market economies had somewhat higher levels of investment - fragile states typically had lower investment levels - Sectoral composition of investment in economic infrastructure (median shares reported): - transportation: about half of total investment in economic infrastructure (usually below 50 percent) - water and sanition: 22 percent - energy: 19 percent - ICT: 6 percent - Observations: - relatively low share of energy investment is troubling given access to electricity as a key constraint to development - private provision of ICT services has allowed governments to spend relatively little in that area ### Private participation in infrastructure (PPPs) - Since 2000 LIDCs accounted for: - 6.5 percent of the value of PPP projects in all emerging market and developing economies - 10.5 percent of the number of PPP projects - PPP flows in LIDCs: - averaged about 0.4 percent of LIDC GDP in the last five years (similar to EMs) - after a sharp acceleration in the early 2010s, PPP flows declined in the most recent years - of the $43 billion in LIDC PPP projects since 2010, more than half has been invested in Asia and one third in Sub-Saharan Africa - Distribution and notable facts: - Vietnam and Bangladesh have the largest number of projects; Lao PDR leads in volume - PPPs have been used for regional projects (example: Central Corridor—integrated transport program across five countries with an investment of about $18 billion) - Project characteristics: - 87 percent greenfield projects since 2000 - 8 percent brownfields - 97 percent of contracts with the central government - variation in project size with very large examples: - coal plant in Laos: $3.7 billion - expansion of the Onne port complex in Nigeria: $2.9 billion - thermal power generation project in Vietnam: $2 billion - nine projects started since 2010 are valued over $1 billion - Role of MDBs: - more than a quarter of projects in LIDCs involve MDB support (direct loans, syndication, equity investment, partial credit guarantees, political risk coverage) - MDB presence is associated with a lower probability that a project comes under distress or is canceled ### Financing for infrastructure: Official Development Finance (ODF) and cross-border lending - Detailed OECD data show: - LIDCs received nearly $17 billion in project finance from MDBs and OECD members in 2014 - 87 percent of ODF for LIDCs consisted of grants and concessional loans (contrast: 56 percent for all developing countries) - bulk of the money went to public projects; direct support to the private sector amounted to $0.9 billion in 2014 - Sectoral shifts in ODF: - share of projects in water and transportation declined steadily since 2006 - share of energy increased to about 30 percent in 2014 - Dispersion across countries: - for all LIDCs in 2014, the median ratio of ODF to GDP equaled 1.3 percent *Source: IMF Working Paper — excerpt "3.1 Public Investment and Saving" (wp17233).* ### 2.0 percent, and the GDP-weighted average 0.9 percent. As expected, grants accounted for ### wp17233 - 2.0 percent, and the GDP-weighted average 0.9 percent. As expected, grants accounted for ### Financing patterns and donor composition - Multilateral support accounted for 57 percent of ODF, bilateral for 43 percent. - ODF commitments amounted to around $24 billion in 2014, exceeding disbursements by a wide margin. - According to OECD (2016), ODF covers 6-7 percent of infrastructure investment in developing countries. - Grants accounted for the bulk of financing in fragile states; frontier markets and commodity exporters received less ODF (relative to their GDP) than other country groups due to a higher domestic revenue base and greater access to commercial borrowing. - In the IMF survey, only 40 percent of LIDC country teams indicated that new projects included a budget for maintenance. ### Role of emerging and non-traditional donors - Between 2000 and 2013 almost 60 percent of Chinese-funded projects were infrastructure ones (AidData). - Gutman et al. (2015) calculate that China contributes about 20 percent of external finance for infrastructure projects in Sub-Saharan Africa, with most of that financing provided by China's EXIM Bank. - India’s development financing for infrastructure is estimated at $1.3 billion in 2014, with most of it going to neighboring countries, primarily for energy and transportation. - New multilateral institutions expanding the donor base include the Asian Infrastructure Investment Bank (AIIB) and the New Development Bank (NDB). ### Cross-border bank lending and commercial flows - MDBs participate in about one fourth of international syndicated cross-border loans to LIDCs. - Vietnam, Uzbekistan, Nigeria, Lao PDR, Ethiopia and Kenya are the largest recipients of international syndicated loans. - Total cross-border bank lending rose steadily in the late 2000s, peaking in 2012—when it amounted to about USD 40 billion—before falling significantly alongside the drop in commodity prices in 2014–15. - Since 2007, almost 30 percent of cross-border bank lending in LIDC financed infrastructure projects; the share in EMs is about 22 percent. - In terms of sector distribution, 52 percent of infrastructure loans go to energy and utilities, 19 percent to telecommunications, 17 percent to transportation. - Cross-border bank lending generally represents a complementary source of external financing with respect to ODF. ### Challenges, absorptive capacity, and efficiency - UNCTAD (2014) estimates that attaining the SDGs would require increasing spending on economic infrastructure by USD 0.8 to 1.7 trillion a year from current levels (covering all developing countries). - Over the last two years, public debt levels have risen, external financing conditions have tightened, and growth prospects have weakened for the LIDCs. - Dabla-Norris et al. (2012) show that low-income countries have relatively weak public investment management institutions; improving those institutions could increase considerably the efficiency (the “value for money”) of public investment. - Isham and Kaufmann (1999) show that once the ratio between public investment over GDP is too high (above 10 percent), the increase in public investment is associated with a declining productivity of investment projects. - Presbitero (2016) shows that infrastructure projects undertaken in periods when public investment accelerates compared to its historical patterns are less likely to be successful, indicating absorptive capacity constraints. - The IMF team survey found multiple obstacles to scaling up public investment in economic infrastructure; no single constraint emerged as dominant in the full sample. Subgroups: fragile state desks emphasized availability of external finance and administrative capacity; frontier economies emphasized availability of domestic resources and limits on debt accumulation. ### Private participation, PPPs, and institutional conditions - Even under optimistic assumptions about future improvements in public investment efficiency, domestic resource mobilization, and concessional financing, addressing the infrastructure shortfall requires a significant increase in private sector participation. - In the near future private participation is likely to occur primarily through PPPs, concentrated in the energy sector. - PPP use is correlated with domestic institutions (rule of law, levels of corruption) and with the Infrascope index; the average Infrascope index for LIDCs is significantly lower than for EMs, with particularly large gaps in the legal regulatory framework and the presence of financial facilities. - Macro-fiscal implications of PPP projects could be large and expose countries to fiscal risks; a strong regulatory environment and a robust institutional framework are essential to implement PPPs sustainably. - There is scope to improve collaboration between local governments and MDBs in the preparation, structuring and financing of infrastructure projects (examples: World Bank Global Infrastructure Facility and the EBRD Equity Participation Fund). ### Policy recommendations and way forward - Mobilize domestic resources for public investment by increasing tax revenue and streamlining and prioritizing expenditures. - Increase access to concessional external financing. - Develop local capital markets. - Strengthen the institutional and regulatory framework to expand private sector involvement in the provision and financing of infrastructure investment, supported by multilateral development banks and development finance institutions. - Improve “value for money” in public and PPP investment projects. - Seek financing on the most concessional terms possible where fiscal space exists, with support from the international community. - Scale up infrastructure investment gradually while building capacity and strengthening institutions to avoid surpassing absorptive capacity limits. ### Conclusions — summary of empirical and policy messages - Public investment, including in infrastructure, has broadly increased in LIDCs over the last 15 years, but the quantity, quality and accessibility of infrastructure remain considerably lower than in emerging market economies. - Outside the telecom sector, infrastructure services in LIDCs are primarily provided by the public sector; private participation is largely channeled through PPPs and has declined recently after an early-2000s spike. - Grants and concessional loans from development partners are essential sources of infrastructure funding in LIDCs; international syndicated loans play an important complementary role in a few countries. - Improving LIDC infrastructure to levels consistent with attaining the Sustainable Development Goals requires coordinated action across public investment efficiency, domestic resource mobilization, concessional financing, institutional strengthening, and a substantial increase in private-sector involvement. *Source: wp17233 (excerpt).* ### Appendix Table A1. LIDCs classification ### Appendix Table A1. LIDCs classification ### Appendix I. Public Investment Scaling-up in Ethiopia - High public investment reflects the government’s national development agenda guided by 5-year Growth and Transformation Plans (GTPs). - Public investment rose from 12 percent of GDP in 2009 to 22 percent in 2015. - Private investment also increased alongside public investment. - Infrastructure outcomes (2010–2015): - Power generating capacity more than doubled. - Number of telecom users quadrupled. - Stock of asphalt roads rose by 30 percent. - Growth of power transmission and distribution networks lagged generation; quality of old lines deteriorated. - Number of electric outages doubled between 2011 and 2015. - Reliance of manufacturing firms on own electricity generators doubled between 2011 and 2015. - Financing for capital spending: - Tax revenue is low even by LIDC standards. - A major compression in current expenditure compared to the 2000s freed up space for public investment. - Debt cancellation under HIPC in the mid-2000s reduced debt service and made room for external borrowing. - External borrowing averaged 5.7 percent of GDP per year over the period 2010-15. - State-owned enterprises (SOEs) have easy access to credit from state-owned banks. - Private banks are required to buy government bonds equivalent to 27 percent of their annual loans to fund long term investments. - Macroeconomic and debt considerations: - Real GDP increased at an average rate of 10 percent per year between 2010 and 2015. - Both domestic and external public debt stood close to 30 percent of GDP in 2015. - Public debt is expected to increase further with implementation of the second GTP. - The 2015 debt sustainability analysis elevated the risk of debt distress from low to medium. - External creditor composition note: - China accounted for 29 percent of total external borrowing during 2012-2015. ### Appendix II. Hydropower PPPs in Lao PDR - Lao PDR has substantial private sector participation in infrastructure projects, especially hydropower, to implement strategy to become “the battery of ASEAN.” - Government uses private investment due to limited ability to invest directly and lack of local finance/expertise; foreign private and public firms and multilateral agencies participate. - PPPs outside energy are limited; two transportation PPPs noted: - $3 million management and lease contract for Vientiane Airport Terminal (successful and still active). - Ngone Bridge Project (initiated 1993, opened 1995, concession later cancelled and operations taken over by government after Asian Financial Crisis in 1997). - Legal/institutional context: - Largely non-existent PPP legal and institutional framework outside energy. - Energy projects use IPP-like structures via locally incorporated limited liability companies with government or state company equity participation. - Typical IPP contractual structure: - Long term concession agreement to exploit a natural resource for energy generation. - Power purchase agreement (PPA). - Engineering, procurement and construction (EPC) contract. - Project financing facilities. - Government participation via Lao State Holding Enterprise (LSHE) which manages concessions and receives concession fees, royalties, and dividends. - Nam Theun 2 (NT2) case study: - NT2 is a $1.45 billion hydroelectric project that began operation in 2010. - NT2 represented the largest foreign investment in Lao PDR, the world’s largest cross-border financing project, and the largest hydroelectric power project in South East Asia. - Concession agreement signed in 2002 between government of Lao PDR and Nam Theun Power Company Limited (NTPC) for 25 years, after which the project is to be transferred to the government. - NTPC shareholders: Electricite de France (EDF) – 35 percent (contracted to carry out construction), Lao State Holding Enterprise (25 percent), Electricity Generating Public Company Ltd. (Thai company – 25 percent), Italian-Thai Development Public Company Limited (Thai entity - 15 percent). - Power Purchasing Agreement: Electricity Generating Authority of Thailand (EGAT) agreed to acquire 95 percent of power produced by NT2 for the first 13 years of operation with pricing indexed to exchange rate and cost of alternative generating technologies; after 13 years energy can be sold on the spot market or likely bought by EDF. - Financing: Equity financing amounted to $450 million; remainder financed by debt from a broad base of lenders including two bilateral lenders, five multilateral lenders, four export credit agencies, and 10 commercial banks; notable contributions from Asian Development Bank, World Bank (IDA) and MIGA. - To date NT2 has provided close to $1 billion of export revenue to Laos and close to $180 million in royalties and dividends to the government. ### Appendix III. Solar Micro-Grids in Kenya - Micro-grids deliver locally produced electricity through low-voltage distribution lines and represent private production and distribution for off-grid communities. - Steama.co platform: - Pioneered the platform in Kenya; micro-grids rapidly expanding and have attracted global players such as E.ON. - Consumers prepay electricity through SMS; consumption measured remotely; payments managed through cloud-based software, reducing administrative costs. - Steama.co started micro-grid operations in 2013 and focuses on providing the management platform to other micro-grid operators. - Initially funded by founders and early investors including Vulcan Capital; in March 2016 Steama.co completed a US $1m seed investment round with “angel investors” led by GReeN and the Ashden Trust. - All figures regarding Steama.co’s operations are provided by Steama.co upon request. - Operational scale and growth (Steama.co platform): - Total installed capacity managed under the Steama.co platform is around 200kW. - Serving 1000 households and businesses with close to 10,000 total end-beneficiaries. - Subscription increased from around 100 connections in 2014 to 1000 in 2015; projected to reach 5000 by end-2016. - Number of service providers increased from 3 in 2014 to 6 in 2016. - Number of solar sites tripled to 31 by 2016. - Generation capacity increased by a factor of 50 between 2014 to 2016. - Comparative costs and affordability: - Price per kWh from the solar micro-grid is around US$1.5. - National grid tariff about US$ 0.15 per kWh. - Levelized cost of electricity (LCOE) in Kenya from individual diesel generators is about US$ 2 per kWh. - Kerosene lanterns cost between US$ 5 to 10 per kWh. - Home solar systems cost between US$ 2.5 to 8 per kWh. - Connection fee to micro-grids is about US$10. - Often subsidized national grid connection fees are US$ 150 to $350; the real cost of a single-phase connection is about US$1000. - Typical micro-grid customer annual spending for lighting, powering television and phone charging is $120. - Average annual lighting spending per household in Kenya is $157. - Market potential and investment trends: - Bloomberg New Energy Finance and Lighting Global (2016) forecast: local solar power will reach one in three off-grid households by 2020. - Off-grid solar (micro-grid and home solar systems) will improve access to electricity for 89 million people in sub-Saharan Africa and Asia by 2020. - Investment in off-grid solar in sub-Saharan Africa increased by 15 fold from $18.4 million in 2012 to $276 million in 2015. - Off-grid population in sub-Saharan Africa and Asia spent over $20.6 billion on lighting in 2014, which is between $45 to $186 per household. - Enabling factors and policy/institutional points: - Synergy between technical advances in electricity generation and mobile telephony (GSM coverage enabling SMS prepayment). - Mobile banking and cloud-based metering/billing reduce administrative burdens and cash-collection risks. - Removal of VAT and tariffs for solar imports has reduced the cost of capital. - Micro-grids are cheaper than off-grid alternatives (diesel, kerosene, home solar systems) and charge much lower connection fees than the national grid, making them attractive in remote rural villages. *Source: wp17233 - Appendix Table A1. LIDCs classification (PDF).* --- _Source: https://www.imf.org/-/media/files/publications/wp/2017/wp17233.pdf_