## cr18176

## Source details

**Canonical URL:** [cr18176](https://www.imf.org/-/media/files/publications/cr/2018/cr18176.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2018/cr18176.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2018/cr18176.pdf.json)

---

### A. Growth Story and TFP Accounting
- Long-run growth and transitions:
  - Sri Lanka has grown at an average of 5 percent over the last four decades.
  - Major transitions: 2009 post-war infrastructure scale-up; 2015 peaceful democratic transition.
- Poverty and inequality (selected figures, 2002–2016):
  - Poverty headcount ratio (US$1.90 a day (2011 PPP)): 8.3 (2002), 3.8 (2006/2007), 2.4 (2009/2010), 1.9 (2012/2013), 0.7 (2016).
  - National poverty lines (percent): 22.7 (2002), 15.3 (2006/2007), 8.9 (2009/2010), 6.7 (2012/2013), 4.1 (2016).
  - Poverty gap (percent): 5.1 (2002), 3.1 (2006/2007), 1.7 (2009/2010), 1.2 (2012/2013), 0.6 (2016).
  - Gini coefficient: 0.40 (2002), 0.40 (2006/2007), 0.36 (2009/2010), 0.39 (2012/2013), 0.45 (2016).
- Aggregate TFP accounting (Solow-residual and utilization-adjusted analyses):
  - Capital was the most important contributor to growth over the past five decades.
  - With opening up in the 70s, TFP contribution reached about 3 percent even during the height of the war and exceeded the average for emerging markets (text table).
  - Capital utilization rates dropped considerably at the peak of the war; labor utilization affected by recurrent natural disasters.
  - Post-2010: capital accumulation picked up due to mega-scale public infrastructure; the TFP residual halved during post-war reconstruction and continued to decline in recent years.
  - FDI new inflows remained stagnant, below 2 percent of GDP since 1998; FDI new inflows dropped to 1.1 percent of GDP in 2016.
- Sensitivity scenarios (selected entries, average growth rate, percent):
  - Solow Residual with Human Capital Adjustment (2010-16): Output 6.3; Employment 0.6; Capital 2.3; Factor Productivity 3.3.
  - Capacity and Labor Utilization and HC Adjustment (2010-16): Output 6.3; Employment 0.5; Capital 3.6; Factor Productivity 2.2.
  - Solow Residual: Depreciation 7% and Capital share 0.4 (2010-16): Output 6.3; Employment 0.7; Capital 3.1; Factor Productivity 2.5.
- Data and methodology notes:
  - Labor share of income taken as 0.7; capital share 0.3.
  - Capital stock estimated with perpetual inventory method and 7 percent depreciation in baseline.
  - Sources: Central Bank of Sri Lanka, Department of Census and Statistics, FAO, Penn World Tables, staff calculations.

### B. Sectoral Decomposition and Structural Change
- Sectoral shares and reallocation:
  - Sectoral shares in output (percent): Agriculture 1960* 26; Manufacturing 1960* 16; All other sectors 1960* 58. By 2016: Agriculture 8; Manufacturing 17; All other sectors 75.
  - Sectoral shares in employment (percent): By 2016 employment shares: Agriculture 25; Manufacturing 17; All other sectors 58.
  - Labor reallocated from lower-productivity agriculture and manufacturing to higher-productivity services and construction; positive contribution of factor reallocation to growth doubled since the war (after an almost three-fold drop during the war).
- Manufacturing performance:
  - Manufacturing TFP has dropped since the 1990s despite growing capital investments.
  - In 2010–2016, capital per worker in manufacturing almost doubled while TFP significantly declined.
  - Largest manufacturing subsectors: food and beverages, textiles, rubber, and chemicals.
- Effective Rate of Protection (ERP) and concentrated protection (2015):
  - ERP for the top 10 most protected manufacturing sectors: range 170–524 percent.
  - Top ten most protected sectors in manufacturing, 2015 (ERP range: 170-524 percent): Processing and preserving of fruit and vegetables; Manufacture of bakery products; Manufacture of macaroni, noodles, couscous and similar farinaceous products; Manufacture of refined petroleum products; Manufacture of refractory products; Distilling, rectifying and blending of spirits & Manufacture of wines; Manufacture of other porcelain and ceramic products; Manufacture of articles of concrete, cement and plaster; Manufacture of soft drinks; production of mineral waters and other bottled waters; Manufacture of dairy products.

### C. Competitiveness, Trade Policy, and Para-tariffs
- Export structure and trade performance:
  - Exports concentrated on garments, tea, and rubber products; share in global trade has been declining gradually.
- Trade policy indicators:
  - IMF trade policy indicators: Sri Lanka’s trade and FDI regimes are more restrictive than the average emerging market in all key areas.
  - World Economic Forum Enabling Trade Index 2016 score: 4.1 out of 7 (main weaknesses: market access, tariff rates and tariffs faced in destination markets).
- Para-tariffs and protectionism:
  - Introduction of para-tariffs during the last decade has effectively doubled protection rates, making the import regime one of the most complex and protectionist in the world.
  - Examples of para-tariffs: Ports and Airports Development Levy (PAL), the Commodity Export Subsidy Scheme (CESS), and the Special Commodity Levy (SCL). The Export Development Board (EDB) Levy was eliminated in 2012.
  - A survey by the International Trade Center covering over 500 businesses identified biggest challenges from para-tariffs: lack of information, extensive documentation, and delays in processing — particularly burdensome for small firms.
- Trade facilitation:
  - Trade facilitation performance measured across 11 WTO Trade Facilitation Agreement dimensions; MFN tariff in IMF indicators does not cover para-tariffs, which are important in Sri Lanka.
- Policy implications (implied by analysis):
  - Simplify tax systems, including trade tariffs and para-tariffs, through predictable, fair, and automated mechanisms to reduce uncertainty and support private-sector-led growth.
  - Stable and transparent regulatory systems to attract long-term foreign investment and support trade integration.
  - Reforms to ease business environment and trade remain at early stages of implementation.

### D. Tax Reform and Investment Framework
- Inland Revenue Act (IRA) reforms:
  - New IRA approved October 2017 moves away from complex tax incentives toward a transparent, unified legal framework.
  - Over past four decades tax expenditures proliferated; recent evidence: "More than one percentage point of GDP, two third of actual revenue collected, is lost to Corporate Income Tax incentives annually."
- FDI outcomes:
  - Sri Lanka’s foreign investment remained below 2 percent of GDP over the past twenty years.
  - Manufacturing received 22-25 percent of total FDIs on average during 2014–2017.
  - Recent new FDI predominantly infrastructure oriented; small proportion reaching global value chain sectors.
- Estimated growth effects of reforms (Lusinyan (2017) illustrative):
  - Product market reforms: increase annual real GDP growth by about 0.6 percentage points.
  - Trade tariff reduction: increase annual real GDP growth by about 0.2 percentage points.
  - Empirical evidence: increased openness raises growth via improvements in TFP and accumulation of physical and human capital.

### E. Infrastructure, Land Access, and Administrative Bottlenecks
- Infrastructure:
  - Sri Lanka has one of the best road and port infrastructures in South Asia after post-war reconstruction; North-East road and transportation networks need further work.
- Land and permits:
  - State ownership of land more than 80 percent; obtaining leases and permits is excessively burdensome due to multiple legislations and parallel land registration systems.
  - Government considering a consolidated land cadaster system and easing lease and permit procedures.

### F. Education, Skills, Labor Markets, and Demographics
- Education and skills indicators:
  - Literacy rate: 93 percent.
  - Only one third of primary school children mastering language and mathematics skills.
  - Only 28 percent of workers can use computers.
  - Only 20 percent of workers are proficient in English.
  - Over 75 per cent of employers expect a high-skill worker to have English language and computer skills.
- Labor market structure and demographics:
  - Overall unemployment 4 percent; youth unemployment 22 percent.
  - Female labor force participation rate: 36 percent.
  - An estimated two-third of workers are employed informally.
  - Working age population expected to grow slowly (13–14 percent rate) and then gradually decline, raising the dependency ratio.
- Policy needs:
  - Build quality of primary and secondary education; expand vocational training and technical skills; introduce soft skills in general education.
  - Enable policies to increase youth employment and female labor force participation.

### G. Gender Gaps, Macroeconomic Costs, and Policy Actions
- Gender disparities and education:
  - Free state education resulted in female school participation rates over 95 percent.
  - Heterogeneity by education level: O-level FLFP about 35 percent (2015); university education FLFP more than 85 percent (2015).
  - World Bank (2018): FLFP shows a skewed-U-shaped curve with respect to education.
- Estimated macro cost of gender gaps:
  - Occupational choice model estimates costs associated with gender gaps in labor force participation and entrepreneurship are currently as high as 20 percent of income per capita compared to equal participation.
- Simulation scenarios (Cuberes, Newiak, Teigner (2017), Sri Lanka-specific):
  - Closing gender gaps in 50 years yields overall income gains of about 16 percent in 2040.
  - Income gains of about 21 percent in 2065 when gender gap completely closed.
- G20 “25 by 25” scenario (cross-country estimates, ILO (2017a)):
  - Increase global labor force by 204 million by 2025 (5.4 percent).
  - Increase global GDP by 3.9 percent or US$5.8 trillion; potential increase in global tax revenue of US$1.5 trillion (emerging countries: US$990 billion; developed countries: US$530 billion).
  - Southern Asia could increase average annual GDP growth by over a percentage point by reaching gender parity by 2025.
- Policies to close gender gaps and GRB:
  - Vision 2025 priorities: affordable child care, improved transport, part-time/flexible work arrangements, improved maternity benefits, improved access to tertiary and vocational training.
  - 2018 budget and cabinet actions: 25 percent local government quota for women (2018 local elections); pilot gender budgeting in twenty ministries; cabinet approval (November 2016) to establish gender mainstreaming in every ministry.
  - Gender-responsive budgeting (GRB): integrate gender analysis across budget cycle; Ministry of Finance should take stronger leadership role.

### H. Fiscal Policy Instruments and Non-Fiscal Measures to Support FLFP
- Fiscal instruments:
  - Use taxes and tax benefits, improved family benefits, subsidized child-care, incentives for businesses to hire women.
  - IRA provisions: basic relief LKR 500,00 and increased employment income relief 700,000 (offset against employment income) can motivate female labor supply by shielding income up to these reliefs from tax.
  - Subsidies and tax incentives can encourage female hiring but fiscal costs must be assessed.
- Expenditure and human capital:
  - Paid parental leave, subsidized child-care, targeted policies for low-income families (examples: US earned income tax credit; UK combinations of tax and transfers).
  - Health: universal free health care reduced mortality, but district disparities persist; drug price regulation and focus on antenatal/postnatal care recommended.
  - Education and skills: expand TVET, strengthen NVQ system, island-wide accreditation of providers.
- PFM tools for gender objectives:
  - Gender budget statements, gender impact assessments, performance-related budgets, gender audits; ensure entry points across budget cycle and that approved policies are funded and implemented.
- Non-fiscal policies:
  - Labor law reform for flexible work, part-time benefits, maternity/paternity leave enforcement, anti-harassment measures.
  - Improve access to finance, especially in plantation and poor areas; include formal financial institutions beyond microfinance.

### I. Disaster Risk Management, Contingency Budgeting, and Electricity Pricing
- Disaster risk profile and impacts:
  - Combined average annual loss estimated at about 0.5 percent of GDP: 0.32 percent for floods; 0.05 percent for droughts; remainder for landslides and cyclones.
  - Large-scale flooding in May 2016 and severe drought late 2016; monsoon May 2017 floods/landslides: >200 lives lost; >9,000 houses damaged.
  - 2016–17 impacts: Agriculture GDP contracted 3.8 percent (2016) and 0.8 percent (2017); food inflation accelerated to 14.4 percent y/y in December 2017; food imports increased by about US$200 million in 2017 (0.2 percent of annual GDP); drought raised oil imports by about US$900 million (1 percent of GDP) in 2017; more than 2 million people affected; food-insecure households estimated at 277,000 in August 2017 (30 percent possibly consuming less than daily minimum calorie intake).
- Three-tier disaster financing approach (by return period):
  - Low-risk layer (about 5 years or less): annual budget allocation/contingency budget; disaster-linked social protection.
  - Medium-risk layer (about 5–20 years): contingent credit for quick post-disaster financing.
  - High-risk layer (>20 years): transfer to international capital/insurance markets (CAT bonds, catastrophe derivatives).
- Contingency budgeting (2017 actual vs 2018 budget, Rs and percent of GDP):
  - 2017 Actual: Total Rs 53.5 bn (0.4 percent of GDP): Mitigation Rs 0.4 bn (0.00 percent of GDP); Rehabilitation Rs 34.4 bn (0.26 percent of GDP); Cash support Rs 5.2 bn (0.04 percent of GDP); Insurance Rs 5.5 bn (0.04 percent of GDP); Other Rs 8.0 bn (0.06 percent of GDP).
  - 2018 Budget: Total Rs 21.0 bn (0.15 percent of GDP): Mitigation Rs 15.2 bn (0.11 percent of GDP); Rehabilitation Rs 0.6 bn (0.00 percent of GDP); Cash support Rs 0.0 bn (0.00 percent of GDP); Insurance Rs 3.5 bn (0.02 percent of GDP); Other Rs 1.7 bn (0.01 percent of GDP).
  - Observations: 2017 rehabilitation and cash support were not appropriated in original budget, requiring mid-year realignment; 2018 budget omits appropriations for rehabilitation and cash support, implying potential future realignments.
- Dedicated contingency fund design features and example (Mexico’s FONDEN): dedicated financing source, governance and investment rules, standing appropriations triggered by events, prudent investment, drawdowns limited to large-impact disasters; FONDEN finances recovery/reconstruction of public assets and low-income houses; FONDEN Trust Fund accumulates unspent disaster budgets.
- Drought and electricity fiscal risk:
  - Hydropower accounts for 30 percent of total electricity generation capacity.
  - 2017 drought raised average electricity cost to about Rs 20 per kWh.
  - End-user tariffs not adjusted since 2014; CEB incurred losses of 0.4 percent of GDP in 2017 due to elevated electricity costs.
  - Policy recommendation: cost-reflective tariff adjustments combined with well-targeted cash transfers; automatic tariff adjustment mechanism under 2015 tariff methodology (cabinet approval expected by September 2018) to minimize subsidy accumulation; protect poor via lifeline tariffs and/or targeted cash transfers.
- Disaster risk transfer instruments (summary):
  - Parametric disaster insurance: fast payouts, data intensive, may be costly.
  - Catastrophe risk pools: economies of scale, price risk, incentivize risk reduction.
  - CAT bonds: principal forgiven on parametric triggers; examples include Mexico and World Bank instruments.
  - Contingent credit example: Cat-DDO effective August 2014, amount $102 million (0.13 percent of 2014 GDP), fully drawn in August 2016 for May 2016 flooding.

### J. Public Insurance Schemes and Improvements
- National Natural Disaster Insurance Scheme (NNDIS):
  - Introduced April 2016; administered by National Insurance Trust Fund (NITF).
  - Coverage: all households and SMEs for property losses up to Rs 2.5 million caused by cyclones, floods, earthquakes (droughts excluded); premiums paid by the government.
  - Reinsurance: NITF purchases reinsurance from international market.
  - Performance: net losses for NNDIS estimated about Rs 2 billion for 2016-17 (0.02 percent of GDP); reinsurance costs about Rs 800 million for 2017; premium paid by government for 2017 Rs 500 million.
- Recommendations to improve NNDIS:
  - Assess risk exposure based on asset values and average payouts.
  - Minimize reinsurance costs by choosing risk layers carefully, adjusting deductibles, and setting appropriate coverage limits.
  - Transition to targeted schemes focusing on poor/vulnerable households; encourage wealthier households to buy private insurance.
  - Use the Social Registry to collect asset data and target poor households consistent with other welfare programs.
  - Improve institutional accounting and transparency (separate accounts for NNDIS and reinsurance).
  - World Bank providing technical assistance.
- Broader disaster insurance improvements:
  - Consider catastrophe insurance for public assets; standardize cover across public assets for scale and diversification benefits.
  - Improve crop insurance schemes, including weather-indexed crop insurance for rice and other crops (2018 Budget measure): contributory scheme with premium paid by both farmers and government.

### K. Financial Sector: Credit Growth and Macroprudential Recommendations
- Credit growth and risks:
  - Significant credit growth for more than two consecutive years warrants attention.
  - Loan concentration in real estate raises concerns.
  - Current macro-financial indicators do not point to systemic risks but low NPLs may understate vulnerabilities.
- Short-term recommendations:
  - Central Bank of Sri Lanka should maintain a prudent data-dependent monetary stance to ensure excessive credit does not drive inflation.
  - Adopt overall macro-prudential framework: loan-to-value limits; higher risk weights on mortgages and construction; calibration to Sri Lanka’s needs.
- Medium-term recommendations:
  - Strengthen supervision and regulation of banking and non-bank financial institutions; improve resolution frameworks.
  - Continue adoption of Basel III practices; improve financial infrastructure and legal framework (e.g., Banking Law Act revisions).

### L. Fiscal Rules, Debt Anchors, and Convergence Scenarios
- Background on FM(R)A:
  - Fiscal Management (Responsibility) Act legislated 2003, revised 2013 and 2016; contains debt, budget deficit, and guarantee rules but weak enforcement mechanisms.
  - Historical outcomes: central government debt fell from around 106 percent (2002) to around 88 percent (2006) but never met original targets; debt ~72 percent in 2013 and nearly 80 percent by 2017; overall deficit never less than 5 percent of GDP from 1990 to 2017.
- Principles for effective fiscal rules (IMF 2018a): sustainability, stabilization, simplicity, operational guidance, resilience, ease of monitoring/enforcement.
- Calibration methodology (debt limit and stochastic simulations):
  - Assumed debt limit for analysis: 70 percent of GDP over medium term (Vision 2025 target).
  - Initial year parameters (values in 2020): Debt-to-GDP 79.5%; Overall budget balance (% GDP) -3.5%; Primary budget balance (% GDP) 2.1%; Real GDP growth 4.7%; Real effective interest rate (domestic) 6.6%; Real effective interest rate (foreign) -0.2%; Foreign debt share 44%.
  - Stochastic-simulation result: a debt anchor of 50 percent of GDP is sufficient to ensure debt stays below the debt limit of 70 percent with 90 percent certainty.
- Convergence scenarios (In percent of GDP unless otherwise noted; initial debt 79.5):
  - Baseline: Overall Balance -3.5; Nominal GDP Growth Rate 10.0; Year Debt Ratio < 70% = 2023; < 60% = 2027; < 50% = 2034.
  - Stronger growth: Overall Balance -3.5; Nominal GDP Growth Rate 12.0; Year Debt Ratio < 70% = 2022; < 60% = 2025; < 50% = 2029.
  - Smaller deficit: Overall Balance -2.0; Nominal GDP Growth Rate 10.0; Year Debt Ratio < 70% = 2023; < 60% = 2025; < 50% = 2029.
  - Weaker growth: Overall Balance -3.5; Nominal GDP Growth Rate 8.0; Year Debt Ratio < 70% = 2025; < 60% = 2033; < 50% = 2052.
  - Larger deficit: Overall Balance -5.0; Nominal GDP Growth Rate 10.0; Year Debt Ratio < 70% = 2026; < 60% = 2037; < 50% = --.
- Policy guidance for fiscal rules:
  - Core recommendation: adopt a second-generation fiscal rule anchored by the debt ratio with the overall balance as operational target (simplicity and operational guidance).
  - Calibration suggests a debt anchor of 50 percent of GDP to keep debt below 70 percent with 90 percent probability.
  - Design features: broad institutional coverage, clear escape clauses, automatic correction mechanisms, enforcement mechanisms, and credible monitoring (dedicated macro-fiscal units, independent fiscal council).
  - Corrective mechanisms should include triggers (quantitative/qualitative), defined correction actions, and timeframes for correction.
- Annex methodological notes:
  - Direct calibration uses multivariate distributions for macro variables and a fiscal reaction function (FRF) estimated across 25 EMs; stock-flow adjustment (SFA) estimated as difference between fitted and actual debt ratios.

### M. Key Policy Recommendations (aggregated)
- Trade, competitiveness, and investment:
  - Review and phase out para-tariffs; conduct comprehensive tariff review; develop time-bound convergence to a simplified tariff structure.
  - Streamline and automate administrative processes for para-tariffs and tariffs.
  - Implement consistent application of the new Inland Revenue Act to provide predictable investment framework.
- Structural and regulatory reforms:
  - Simplify tax incentives; reduce poorly targeted tax expenditures that cost "More than one percentage point of GDP" annually.
  - Strengthen business environment reform to attract FDI beyond infrastructure.
  - Consolidate land administration (land cadaster) and ease lease/permit procedures.
- Human capital and labor markets:
  - Strengthen education quality, vocational training, and soft skills; boost computer and English proficiency.
  - Promote policies to increase female labor force participation and youth employment (child care, transport, flexible work, GRB).
  - Implement gender-responsive budgeting across ministries with Ministry of Finance leadership.
- Disaster risk and fiscal resilience:
  - Adopt contingency budgeting for emergency cash support and infrastructure rehabilitation within total spending envelope and consistent with fiscal targets.
  - Consider medium-term natural disaster reserve fund; build capacity for parametric insurance, catastrophe pools, and CAT bonds.
  - Introduce automatic electricity tariff adjustment mechanism and protect poor via lifeline tariffs or targeted cash transfers.
  - Improve NNDIS risk management and consider targeted coverage for vulnerable households; develop weather-indexed crop insurance under contributory schemes.
- Fiscal rules and public debt:
  - Adopt debt-anchored fiscal rule with overall balance operational target; calibrate anchor to 50 percent of GDP.
  - Define escape clauses, automatic correction mechanisms, stronger monitoring, and independent fiscal oversight.
- Financial stability:
  - Maintain prudent data-dependent monetary stance amid credit growth; adopt macroprudential measures (LTV limits, higher risk weights for mortgages/construction).
  - Strengthen supervision, resolution frameworks, and financial infrastructure.

*Central Bank of Sri Lanka, Department of Census and Statistics, Food and Agriculture Organization, Penn World Tables, and staff calculations.*

### References _____________________________________________________________________________ 24

### SRI LANKA: STRUCTURAL TRANSFORMATION—THE NEW FRONTIER

### A. Growth Story
- Sri Lanka has grown at an average of 5 percent over the last four decades.
- Major transitions:
  - 2009: post-war scale up of infrastructure.
  - 2015: peaceful and democratic transition of power to the coalition government.
- Policy orientation history: state control and import substitution in early 70s; two waves of liberalization in early 80s and 90s; closing up in early 2000s; opening up again since the end of the war.
- Poverty and inequality highlights (selected figures from 2002–2016):
  - Poverty headcount ratio (US$1.90 a day (2011 PPP)): 8.3 (2002), 3.8 (2006/2007), 2.4 (2009/2010), 1.9 (2012/2013), 0.7 (2016)
  - National poverty lines (percent): 22.7 (2002), 15.3 (2006/2007), 8.9 (2009/2010), 6.7 (2012/2013), 4.1 (2016)
  - Poverty gap (percent): 5.1 (2002), 3.1 (2006/2007), 1.7 (2009/2010), 1.2 (2012/2013), 0.6 (2016)
  - Gini coefficient: 0.40 (2002), 0.40 (2006/2007), 0.36 (2009/2010), 0.39 (2012/2013), 0.45 (2016)
- Authors/preparers: Eteri Kvintradze, Amitha Sundararaj, and Misa Takebe.

### Aggregate TFP Accounting
- Simple Solow residual growth accounting findings:
  - Capital was the most important contributor to growth over the past five decades.
  - With opening up in the 70s, TFP contribution reached about 3 percent even during the height of the war and exceeded the average for emerging markets (text table).
- Factor utilization effects:
  - Capital utilization rates dropped considerably at the peak of the war.
  - Labor utilization rates possibly affected by recurrent natural disasters (Annex I).
- Post-war period (post-2010) observations:
  - Capital accumulation picked up driven by mega-scale public infrastructure development projects.
  - The TFP residual halved during the post-war reconstruction period and continued to decline in recent years.
  - FDI new inflows remained stagnant, below 2 percent of GDP since 1998; FDI new inflows dropped to 1.1 percent of GDP in 2016.
- Sectoral contributions to output and productivity (selected average growth rates, 1961–2016 and subperiods):
  - Output, employment, capital, output per worker, capital per worker, factor productivity reported in tabulated form for periods including 1961-70, 1971-77, 1978-87, 1988-02, 2003-09, 2010-16 (see tables in source).

### Sectoral Decomposition and Factor Reallocation
- Structural change:
  - Agriculture share in output and employment declined over the years; services, construction, and other sectors became main employers and output drivers.
- Sectoral shares in output and employment (selected years):
  - Sectoral shares in output (percent): Agriculture 1960* 26; Manufacturing 1960* 16; All other sectors 1960* 58 (other sectors = construction, quarrying and mining, and services). By 2016: Agriculture 8; Manufacturing 17; All other sectors 75 (output).
  - Sectoral shares in employment (percent): 1963 (employment baseline) and later years shown in tables; by 2016 employment shares: Agriculture 25; Manufacturing 17; All other sectors 58.
- Manufacturing sector performance:
  - Manufacturing TFP has dropped since the 1990s despite growing capital investments.
  - In 2010–2016, capital per worker in manufacturing almost doubled while TFP significantly declined.
  - Largest manufacturing subsectors: food and beverages, textiles, rubber, and chemicals.
- Reallocation effects:
  - Labor moved from lower-productivity agriculture and manufacturing to higher-productivity services and construction.
  - The positive contribution of factor reallocation to growth doubled since the war (after an almost three-fold drop during the war).
- Sectoral growth accounting (adjusted for utilization) — average growth rates for 1990-2002, 2003-2009, 2010-2016 show contributions to growth from output, employment, capital, output per worker, capital per worker, and factor productivity by sector (agriculture, manufacturing, all other sectors).

### B. Competitiveness Challenges and Opportunities
- Export structure and trade performance:
  - Sri Lanka’s export structure remains concentrated on garments, tea, and rubber products (Annex II).
  - Sri Lanka has not diversified exports; share in global trade has been declining gradually.
- Trade policy and competitiveness indicators:
  - According to IMF-developed trade policy indicators, Sri Lanka’s trade and FDI regimes are more restrictive than the average emerging market in all key areas.
  - Sri Lanka is more distant from the emerging market average in trade facilitation performance and ease of starting a foreign business.
  - World Economic Forum’s Enabling Trade Index 2016 score: 4.1 out of 7 (main weaknesses: market access, tariff rates and tariffs faced in destination markets).
- Para-tariffs and protectionism:
  - Introduction of para-tariffs during the last decade has effectively doubled protection rates, making the present import regime one of the most complex and protectionist in the world.
  - Para-tariffs increase both monetary and procedural costs for firms.
  - Examples of para-tariffs: Ports and Airports Development Levy (PAL), the Commodity Export Subsidy Scheme (CESS), and the Special Commodity Levy (SCL). The Export Development Board (EDB) Levy was eliminated in 2012.
  - A survey by the International Trade Center covering over 500 businesses found the biggest challenges from para-tariffs are lack of information, extensive documentation, and delays in processing — particularly burdensome for small firms.
- Trade facilitation detail:
  - Trade facilitation performance is measured as the average of 11 dimensions of the WTO’s Trade Facilitation Agreement (information availability, appeal procedures, fees and charges, documentation formalities, etc.).
  - The MFN tariff in the IMF indicators does not cover para-tariffs, which play an important role in Sri Lanka.
- Competitiveness policy implications (implied by analysis):
  - Simplifying tax systems, including trade tariffs and para-tariffs, through predictable, fair, and automated mechanisms can remove uncertainties that hold back trade and investment decisions and support a shift to private-sector-led growth.
  - Stable and transparent regulatory systems would make Sri Lanka’s business environment more attractive for long-term foreign investment and support trade integration.
  - Reforms to ease the business environment and trade are still at early stages of implementation; business environment liberalization is an important catalyst of private-sector-led growth.

*Italic source attribution line.*

### 14.      The most protected segments of the economy benefit from both high tariffs and para-

### 14.      The most protected segments of the economy benefit from both high tariffs and para-tariffs

### Effective protection and concentrated sectoral protection
- The Effective Rate of Protection (ERP) for the top 10 most protected sectors reach between 170 and 524 percent as of 2015.
- Protection is granted by tariff, non-tariff, and/or tax exemptions.
- Some sub-sectors have a third layer of protection on their inputs through selected tax holidays or exemptions.
- Some of the most protected sectors do not necessarily represent strategic development or high employment activities, but appear effective in lobbying for protection.
- The increase in protective barriers coincides with the decline in total contribution of state-owned establishments to sectoral value added; in some sectors the government gave up direct ownership but ensured lasting protection of private companies.

Top Ten Most Protected Sectors in Manufacturing, 2015 (ERP range: 170-524 percent)
- 1 Processing and preserving of fruit and vegetables
- 2 Manufacture of bakery products
- 3 Manufacture of macaroni, noodles, couscous and similar farinaceous products
- 4 Manufacture of refined petroleum products
- 5 Manufacture of refractory products
- 6 Distilling, rectifying and blending of spirits & Manufacture of wines
- 7 Manufacture of other porcelain and ceramic products
- 8 Manufacture of articles of concrete, cement and plaster
- 9 Manufacture of soft drinks; production of mineral waters and other bottled waters
- 10 Manufacture of dairy products

### Policy commitments and trade reform actions
- As part of the Vision 2025 priorities, the government announced it will adopt a time-bound plan to eliminate para-tariffs and conduct a comprehensive review of tariffs.
- The government committed to eliminate 1200 lines of para-tariffs in the 2018 budget and is developing a 3-year plan to fully phase out remaining para-tariffs through subsequent budgets.
- Policy discussions ongoing on a comprehensive review of tariffs and developing a time-bound plan for convergence to a relatively simplified structure.
- Streamlining administrative processes and simplifying para-tariffs and tariffs are expected to have a beneficial impact on trade.
- Under the World Bank trade and competitiveness program (launched March 2017), key achievements include:
  - Improving the investment climate: Investment Climate Reform Road Map and Action Plans; Investment Promotion Certification program completed in January 2018; switch from tax holidays to investment allowances in April 2018 with the new Inland Revenue Act.
  - Eliminating barriers to trade: trade vision document adopted; National Trade Facilitation Committee Secretariat operational; National Trade Facilitation Action Plan finalized.
  - Fostering enterprise competitiveness: Tourism Vision 2025 and Tourism Strategic Action Plan approved; Innovation and Entrepreneurship Roadmap and related strategies adopted.

### Tax reform and investment framework
- The new Inland Revenue Act (IRA), approved October 2017, moves away from complex and volatile tax incentives toward a transparent, unified, and even-handed legal framework to guide long-term investment decisions.
- Over the past four decades tax expenditures were used extensively and rarely repealed, leading to proliferation of tax expenditures without necessarily delivering on development goals.
- Recent studies show incentives benefit a large portion of firms which would be profitable even without incentives.
- Tax incentives were not well targeted and costly in terms of forgone revenue: "More than one percentage point of GDP, two third of actual revenue collected, is lost to Corporate Income Tax incentives annually."

### Competitiveness, FDI, and structural reform impact
- Sri Lanka’s foreign investment has remained below 2 percent of GDP over the past twenty years.
- On average, manufacturing sector received 22-25 percent of total FDIs during the period 2014–2017.
- New FDI in the past few years has been predominantly infrastructure oriented with only a relatively small proportion reaching sectors associated with global value chains.
- Structural reforms illustrative impact on annual real GDP growth (to reach average emerging market economies over a ten-year period, based on Lusinyan (2017)):
  - Product market reforms: increase by about 0.6 percentage points per year.
  - Trade tariff reduction: increase by about 0.2 percentage points per year.
- Empirical evidence (Weerasinghe (2005)) finds an increase in openness raises growth via improvements in total factor productivity and accumulation of physical and human capital.

### Infrastructure, land access, and administrative bottlenecks
- Sri Lanka has one of the best road and port infrastructure in the South Asian region after post-war reconstruction, but more remains to be done for North-East road and transportation networks.
- State ownership of land is more than 80 percent; obtaining leases and permits is excessively burdensome with multiple legislations and parallel land registration systems.
- The government is considering a consolidated land cadaster system and easing lease and permit procedures.

### Education, skills, and labor market challenges
- Education and skills indicators:
  - Literacy rate: 93 percent.
  - Only one third of primary school children mastering language and mathematics skills.
  - Only 28 percent of workers can use computers.
  - Only 20 percent of workers are proficient in English.
  - Over 75 per cent of employers expect a high-skill worker to have English language and computer skills.
- Labor market structure and demographics:
  - Low unemployment rate masks segmentation: overall unemployment 4 percent; youth unemployment 22 percent.
  - Female labor force participation rate: 36 percent.
  - An estimated two-third of workers are employed informally.
  - Working age population expected to grow slowly (13–14 percent rate) and then gradually decline, raising the dependency ratio.
- Policy needs: build quality of primary and secondary education, expand vocational training and technical skills, introduce soft skills in general education, and enable policies to increase youth employment and female labor force participation.

### Conclusions and policy recommendations
- The current trade regime, including para-tariffs introduced during the last decade, has effectively doubled the protection rate, making it highly complex and protectionist.
- Despite extensive tax incentive systems, FDI remains low; reforms to create a stable and transparent regulatory system are critical to attract long-term investment.
- Priority policy actions highlighted:
  - Review and phase out para-tariffs and conduct a comprehensive review of tariffs with a time-bound convergence to a simplified structure.
  - Streamline administrative processes and automate approval systems for para-tariffs and tariffs.
  - Consistent implementation of the new IRA to provide a predictable investment framework.
  - Consolidate land administration (land cadaster) and ease lease and permit procedures.
  - Strengthen education quality, vocational training, and soft skills; boost computer and English proficiency.
  - Promote policies to increase female labor force participation and youth employment to mitigate demographic pressures.
  - Continue open consultative reform processes to build public consensus.

*Source: Department of Census and Statistics and Staff Calculations*

### Annex I. TFP Accounting and Data Sources

### Annex I. TFP Accounting and Data Sources

### Methodologies
- Three aggregate TFP accounting methods summarized:
  - Solow Residual TFP
    - Production function specification and differentiation presented (Solow residual formulation preserved).
    - Equation form shown in source (variables: output ܻ௧, TFP ܣ௧, capital ܭ௧, human capital and utilization adjustments where applicable, α output elasticity).
  - Human Capital Adjusted TFP
    - Human capital adjustment denoted by ݄௧; utilization-adjusted labor is product of labor utilization ܷ௧ and persons engaged ܮ௧.
    - Growth-accounting expression includes human capital adjustment terms.
  - Capital and Labor Utilization Adjusted TFP
    - Incorporates capital utilization ݖ௧ and labor utilization ܷ௧ in the production function and resulting growth accounting identity.
- General definitions and notation:
  - Output ܻ௧ is a function of ܣ௧ (TFP), the stock of physical capital ܭ௧, and utilization ݖ௧.
  - Human capital adjustment: ݄௧.
  - Utilization-adjusted labor: ܷ௧ × ܮ௧ (persons engaged).
  - α is the output elasticity of physical capital; 1- α is the output elasticity of augmented labor.
- Sectoral decomposition:
  - Uses Bosworth and Collins (2006) approach to show reallocation effects in growth accounts.
  - Sectoral decomposition separates sectoral reallocation effects from sectoral composition effects using employment shares and value-added shares by sector.

### Data Definition and Sources
- General note:
  - Accuracy depends on methodology and measurement of factor inputs and factor shares.
  - Data inconsistencies and quality limitations are acknowledged (Weerasinghe 2005).
- Output:
  - Gross domestic production, real growth rates, and sectoral contribution: obtained from the Special Appendix Table of Annual Report 2016 and Economic Social Statistics 2017 by Central Bank of Sri Lanka (CBSL).
  - GDP deflator (2010=100): taken from World Development Indicators.
  - Sectoral real growth rates: year-on-year change in gross domestic production/gross value added in constant prices with different base years; these growth rates used to extrapolate sectoral and aggregate real GDP in 2010 prices.
  - National Account estimates revised in 1970, 1982, and 1996 by CBSL and Department of Census and Statistics in 2002 and 2010; consistent long series not readily available and therefore estimated using real growth rates of different base years in 2010 prices.
- Capital:
  - Gross fixed capital formation in constant local currency, inventories in constant local currency, and adjusted savings: consumption of fixed capital (% of GNI) obtained from World Development Indicators.
  - Capital stock estimated under the perpetual inventory method with 7 percent depreciation rate using gross fixed capital formation and changes in inventories.
- Labor:
  - Number of persons engaged, share of labor compensation in GDP, and annual average hours worked: taken from Penn World Tables.
  - Sectoral labor shares as percent of total employment: taken from labor force surveys (1990 onwards) and similar surveys (before 1990) from Department of Census and Statistics obtained from CBSL Annual Report 2000 (Special Statistical Appendix – Table 4).
  - Missing values imputed using fixed growth assumption.
  - Pre-1990 labor data sources: Census of Population, Socio Economic Survey, Land and Labor Utilization Survey (irregular intervals).
  - Primary source of labor force and employment from 1990: Labor Force Survey by Department of Census and Statistics (DCS); two provinces were inaccessible during conflict leading to anomalies in national data. Persons engaged from Penn World Tables were considered in place of national series where appropriate.
- Utilization Rates:
  - CBSL publishes capacity utilization series in annual reports based on factory industrial production surveys on output and usage of installed capacity.
  - Capacity utilization rates for 1968–1975: taken from CBSL Annual Report 2000, Special Appendix Table 14, extracted from "The Economy of Sri Lanka".
  - Labor utilization rates:
    - Estimated using average annual hours worked by persons engaged from Penn World Tables (PWT) until 2014.
    - CBSL’s new series on gross value added per hour worked used since 2014.
    - Maximum workable hours in a week assumed to be 40.
- Labor and Capital Share of Income:
  - Cobb-Douglas production function assumed with constant returns to scale.
  - Labor share of income taken as 0.7, a 15-year average based on Penn World Table data for Sri Lanka.
  - Capital share is therefore 0.3.
  - Note: literature shows capital share varies between 0.3-0.4 for emerging market countries; sensitivity analysis presented for alternative values and depreciation rates.
- Human Capital Index and Literacy Rate:
  - Human Capital Index for Sri Lanka from Penn World Table 9.0 is based on average years of schooling from Barro and Lee (BL, 2013) and an assumed rate of return to education (Mincer equation estimates, Psacharopoulos, 1994).
  - Free education policy implemented in the late 1945.
  - Literacy rate: 1911 = 31%; 1971 = 78.1%; high human capital development during early 1970s consistent with prevailing policies.
- Agriculture:
  - FAO Agriculture Capital Stock (ACS) database provides country-by-country GFCF for agriculture, forestry and fishing since 1990.
  - For this study: GFCF in current local currency for agriculture deflated by deflator (2010=100) taken without adjustments.
  - Perpetual inventory method with 7 percent depreciation used to estimate capital stock.
  - Depreciation rate aligns with consumption of fixed capital as percent of agriculture value added in FAO database.
  - Land omitted as input following Bosworth and Collins (2007) observation that including land does not substantially change results.
- Industry, Manufacturing:
  - Employment and gross additions to fixed capital formation for manufacturing and industries (without construction activities): taken from Annual Survey of Industries by Department of Census and Statistics.
  - Missing values and 2015–16 values for gross fixed capital formation calculated using the total capital stock growth rate.
- Effective Rate of Protection (ERP):
  - Derived from Input–Output Tables published by Department of Census and Statistics.
  - Department calculated nominal and effective rates of protection for 127 industries using nominal tariff rates, input shares from Supply-Use Table and Symmetric Input Output Table (SIOT) framework for Sri Lanka for 2010 and 2015.
  - I/O table approach covers all inputs and outputs and accounts for usage of tradables (complete and consistent).
  - ERP expression from source preserved: ܴܲܧ݆ = ݐ݆ െ (∑݆ܽ݅ ݐ݅)/ (∑݆ܽ݅) with definitions:
    - ݐ௝ - nominal tariff on j,
    - ݐ௜ - nominal tariff on i,
    - ܽ௜௝ - share of final value of j accounted for by input i.

### Sensitivity Analysis (summary of findings)
- Using 2002 base year growth rates for period 2010–2015, TFP growth rates increase by 0.4–0.5 percent range across different methodologies.
- Selected results from sensitivity scenarios (average growth rate, in percent). Contribution components: Output, Employment, Capital, Factor Productivity. (Table entries preserved as in source.)
  - Solow Residual with Human Capital (HC) Adjustment
    - 1961-70: 4.7 4.3 2.8 -2.5
    - 1971-77: 2.9 2.4 2.7 -2.2
    - 1978-87: 5.4 4.2 3.3 -2.1
    - 1988-02: 4.6 3.1 1.4 0.1
    - 2003-09: 6.0 1.2 1.6 3.1
    - 2010-16: 6.3 0.6 2.3 3.3
  - Capacity Utilization and HC Adjustment
    - 1971-77: 2.9 2.4 2.3 -1.8
    - 1978-87: 5.4 4.2 4.3 -3.0
    - 1988-02: 4.6 3.1 1.4 0.1
    - 2003-09: 6.0 1.2 0.5 4.2
    - 2010-16: 6.3 0.6 3.6 2.0
  - Capacity and Labor Utilization and HC Adjustment
    - 1971-77: 2.9 4.8 2.3 -4.2
    - 1978-87: 5.4 4.1 4.3 -3.0
    - 1988-02: 4.6 3.1 1.4 0.1
    - 2003-09: 6.0 1.4 0.5 4.0
    - 2010-16: 6.3 0.5 3.6 2.2
  - Solow Residual: Depreciation 5% and Capital share 0.3
    - 1961-70: 4.7 1.8 2.7 0.2
    - 1971-77: 2.9 1.2 2.6 -0.9
    - 1978-87: 5.4 2.0 3.2 0.1
    - 1988-02: 4.6 0.8 1.4 2.4
    - 2003-09: 6.0 1.1 1.6 3.2
    - 2010-16: 6.3 0.8 2.3 3.2
  - Utilization Adjusted: Depreciation 5% and Capital share 0.3
    - 1971-77: 2.9 3.7 2.2 -3.0
    - 1978-87: 5.4 2.0 4.2 -0.8
    - 1988-02: 4.6 0.7 1.5 2.4
    - 2003-09: 6.0 1.3 0.5 4.1
    - 2010-16: 6.3 0.6 3.6 2.1
  - Solow Residual: Depreciation 7% and Capital share 0.4
    - 1961-70: 4.7 1.5 3.8 -0.6
    - 1971-77: 2.9 1.0 3.6 -1.7
    - 1978-87: 5.4 1.7 4.4 -0.7
    - 1988-02: 4.6 0.7 1.8 2.1
    - 2003-09: 6.0 1.0 2.1 2.8
    - 2010-16: 6.3 0.7 3.1 2.5
  - Utilization Adjusted: Depreciation 7% and Capital share 0.4
    - 1971-77: 2.9 3.1 3.1 -3.3
    - 1978-87: 5.4 1.7 5.7 -2.0
    - 1988-02: 4.6 0.6 1.9 2.1
    - 2003-09: 6.0 1.1 0.7 4.1
    - 2010-16: 6.3 0.5 4.8 1.0
- Source attribution for data in sensitivity analysis: Central Bank of Sri Lanka, Department of Census and Statistics, Food and Agriculture Organization, Penn World Tables, and staff calculations.

*Italicized source attribution line as in original: Central Bank of Sri Lanka, Department of Census and Statistics, Food and Agriculture Organization, Penn World Tables, and staff calculations.*

### 2.      Sri Lanka lags behind in gender parity in the labor force compared to its emerging

### 2.      Sri Lanka lags behind in gender parity in the labor force compared to its emerging

### Sri Lanka's gender gaps and link between education and economic participation
- Provision of free state education with incentives (free textbooks, free uniforms, scholarships, subsidized transport for all, and free school meals for the children of the economically disadvantaged) has raised female participation rates at school to over 95 percent.
- Sri Lanka ranks below all ASEAN countries and several low-income developing countries in terms of economic opportunity for women.
- Heterogeneity by education level:
  - O-level education: 2015 FLFP rate is about 35 percent.
  - University education: 2015 FLFP rate increased sharply to more than 85 percent (as opposed to less than 80 percent in 2009).
- World Bank (2018) describes FLFP as showing a skewed-U-shaped curve with respect to education.
- Notes on terminology: O-level and A-level refer to education up to 10 and 12 years of schooling.

### Macroeconomic costs of gender gaps and occupational misallocation
- Literature and models indicate significant macroeconomic gains from improving gender parity; country-specific simulations for Sri Lanka show substantial potential gains.
- Occupational choice model (Cuberes and Teigner, 2014):
  - Frictions on economic participation and wages of women (reflecting discrimination, differences in choices, or other demand/supply factors) prevent optimal occupational choices and reduce per capita income.
  - Model agents choose as employers, self-employed, or employees; female frictions distort optimal allocation.
- Estimated current costs for Sri Lanka:
  - Costs associated with gender gaps in labor force participation and entrepreneurship are currently as high as 20 percent of income per capita compared to a situation where women have the same level of labor force and entrepreneurship participation as men.
- Demographics and mitigation:
  - Sri Lanka’s dependency ratio (non-working-age population to working-age population ages 15–64) has already started to rise.
  - Integrating educated women (especially those with an A-level degree or lower) could reverse negative effects of declining labor share on real GDP per capita growth.

### Simulation scenarios and projected income gains
- Cuberes, Newiak, and Teigner (2017) scenarios for declining labor force due to rising dependency ratio examine four paths for closing gender gaps:
  1. No change in gender gaps.
  2. Constant decrease with elimination in 50 years.
  3. Constant decrease with elimination in 100 years.
  4. Constant decrease with elimination in 150 years.
- Sri Lanka-specific simulation results:
  - Overall income gains of about 16 percent in 2040 if gender gaps are closed in 50 years.
  - Income gains of about 21 percent in 2065 when the gender gap is completely closed.
  - Faster declines in gender gaps would yield higher gains.
- Assumptions underlying these results include: increases in FLFP do not negatively affect fertility rates, male participation rates, or human capital accumulation.

### Cross-country econometric estimates and the “25 by 25” scenario
- G20 “25 by 25” target: reduce the gap in participation rates between men and women by 25 percent by 2025.
- ILO (2017a) estimates applying “25 by 25” globally:
  - Increase global labor force by 204 million by 2025, equivalent to an increase of 5.4 percent.
  - Boost global employment by 189 million (or 5.3 percent), holding unemployment rates and male participation rates constant.
  - Increase global GDP by 3.9 percent or US$5.8 trillion.
  - Potential increase in global tax revenue of US$1.5 trillion given currently projected government revenue shares in GDP.
    - Of this, emerging countries: US$990 billion.
    - Developed countries: US$530 billion.
- Regional implication for Southern Asia:
  - Countries in Southern Asia could increase their average annual GDP growth by over a percentage point by reaching gender parity by 2025.
  - Southern Asia average FLFP is estimated to be about 29 percent, below Sri Lanka’s 36 percent rate.
  - Regional analysis suggests Sri Lanka could increase GDP around the same magnitude as the regional estimate.
- Assumptions for the “25 by 25” scenario:
  - Female participation rate increases such that the gap between male and female participation rate decreases by 25 percent by 2025.
  - Each additional woman succeeds in finding a job at the same rate as current women in the labor market (female unemployment rate remains unchanged).
  - Additional women attain 80 percent of the average labor productivity in the respective year and country.

### Policies to close the gender gap and gender-responsive budgeting (GRB)
- Authorities’ strategies and short-term measures:
  - Vision 2025 outlines medium-term strategy to:
    - Provide affordable child care facilities.
    - Improve access to transportation.
    - Facilitate part-time and flexible work arrangements.
    - Improve maternity benefits for private sector employees.
    - Improve access to tertiary education and vocational training.
  - Authorities taking steps to increase female political representation (2018 local elections had allocated 25 percent quota to women in local government).
  - 2018 budget includes several proposals to support gender equality in the labor force.
  - Cabinet approved a strategy to invest at least 25 percent of the provision on state projects for uplifting of rural economy for women and ensure their participation in these projects.
- Gender-responsive budgeting (GRB) framework:
  - GRB uses fiscal policy and public financial management (PFM) to promote gender equality.
  - GRB involves: fiscal policies related to gender objectives and gender-responsive PFM practices across the budget cycle (preparation, allocation, prioritization, execution, monitoring and evaluation).
  - GRB is an analysis of the entire budget process through a gender lens to identify gender-differential impacts and translate gender commitments into budgetary commitments; it is not primarily an issue of additional resources.
- Sri Lanka’s GRB history and planned implementation:
  - Sri Lanka joined the Commonwealth’s gender budgeting pilot project in 1997.
  - Past budgets mentioned reducing the gender gap but lacked systematic attention to FLFP and proper implementing or monitoring mechanisms.
  - 2018 budget commits to supporting FLFP, but mapping specific measures to stated outcomes would be more effective.
  - Authorities are considering implementing gender budgeting in twenty ministries on a pilot basis.
  - November 2016 Cabinet approval to establish gender mainstreaming programs in every ministry, implemented under the supervision of the Additional Secretary/Director of each ministry.
  - Authorities have mobilized gender focal points in ministries and provincial councils; necessary training for focal points is required.
  - The Ministry of Finance should play a stronger leadership role in institutionalizing gender within the conceptual and decision-making framework of budgeting and PFM.

*Source: cr18176 - 2.      Sri Lanka lags behind in gender parity in the labor force compared to its emerging*

### 16.      In this context, some of the policy instruments can be considered as follows:

### cr18176 - 16.      In this context, some of the policy instruments can be considered as follows:

### I. Fiscal policy instruments
- Fiscal policy instruments relevant to gain gender equality include:
  - the use of tax and tax benefits to increase the supply of female labor,
  - improved family benefits,
  - subsidized child-care,
  - other social benefits that increase the net return to women’s work,
  - incentives for businesses to encourage the hiring of women (IMF, 2017a,b).
- Implementation of adequate fiscal reforms can help Sri Lanka increase its fiscal space so finances can be channeled towards social spending in education, health and social safety net to support greater female labor market participation (FLFP) and inclusive growth.
- Relevance is heightened by the shrinking demographic dividend with ageing populations and higher old-age dependency ratios.

- Subsidies and tax incentives:
  - Public subsidies and tax incentives can encourage businesses to hire women, including from underrepresented or minority groups, but fiscal costs must be assessed.
  - The new IRA has a clean framework that does not specify any specific tax policy positions directed towards FLFP.
  - The new IRA has both a basic relief (in the amount of LKR 500,00 for each year) and an increased employment income relief (in the amount of 700,000) that can be offset against employment income, which can be a motivation to females to join the labor force as there would be no tax payable on income up to this level of combined relief.

- General expenditure policies:
  - Policies to support female labor supply include paid parental leave, subsidized child-care, and other social benefits that increase the net return to women’s work.
  - Policies can target low-income families with gender perspective; examples cited: earned income tax credit (US), combinations of tax and transfers (UK and other G7 countries).

- Building human capital through better health:
  - Universal access to free health care services for seven decades has reduced mortality rates, especially among women, but district-wide disparities remain high, with highest mortality in plantation and North-east districts.
  - Utilization of antenatal and postnatal care, immunization, and institutional births are near universal; however rising cost of drugs in an unregulated market creates hardships for the poor.
  - Greater focus on these areas is needed (ILO, 2016; Solotaroff et.al., and World Bank, 2013).

- Closing skills mismatch through education:
  - Girls have higher enrollment and retention in secondary education and higher performance at public examinations; higher percentages of female students in universities and non-vocational tertiary education versus male students.
  - Wide gender imbalances in enrollment in technological courses (including information technology) and technical-related vocational courses limit access to high skilled, remunerative technology-related employment.
  - Recommended: expand technical and vocational education and training (TVET) programs, strengthen the National Vocational Qualification system, and island-wide accreditation of providers to boost skill development (ADB, 2015; ILO, 2015, 2016 and 2017b; World Bank, 2013).

- Fair and equivalent coverage under social insurance:
  - Reforms of social security and pensions should account for women’s intermittent work history, greater incidence of part-time work, lower earnings, and need for parental leave.

- Child-care provision:
  - Vision 2025 identifies improving access to good quality and affordable child care facilities as key to facilitate FLFP.
  - Revenue based fiscal consolidation can make space for government initiatives to support affordable child care schemes.
  - Studies show childcare subsidies can be broadly budget neutral as higher income tax payments from women compensate subsidy cost (Kolovich et.al, 2017).
  - Country examples:
    - Egyptian authorities increased budget resources to public nurseries in 2016 and are developing an integral strategy to improve child care quality.
    - Canada amended child care benefit regime in 2016 with the new Canada Child Benefit to increase support to low and middle-income families (compared to previous universal benefit).
    - UK in 2015 increased childcare support for low-income working parents by providing tax allowance for childcare.
  - In Sri Lanka: authorities are considering an easy credit system with banks for private entrepreneurs to run day care centers and redefining target-group selection for state subsidies to save funds for child care benefits.

- Safe transportation:
  - Vision 2025 identifies transportation as an element to boost FLFP; improvements in safe transportation for women are critical to lower gender based violence prevalent on transportation and in public spaces.
  - Examples:
    - Mexico City’s Viajemos Seguras program and women-only public transport buses.
    - Kerala’s 2017–18 Gender and Child Budgeting plan targets (1) skill development, employment generation, and livelihood security for vulnerable women, and (2) preventing violence against women; budget allocations planned for entrepreneurship, skills training, child care, and gender-friendly infrastructure (Kolovich and Loungani, 2018).
  - Note: Sri Lanka has a National Action Plan to Address Sexual and Gender Based Violence (2016-2020) however stronger implementation of specific measures is needed including in providing safe transportation.12

### II. Public Financial Management (PFM)
- PFM tools to promote FLFP include:
  - gender budget statements,
  - gender impact assessments,
  - performance-related budget frameworks,
  - gender audits.
- The Ministry of Finance plays a key role in promoting and coordinating gender budgeting and associated analytical tools.
- PFM institutions enable operationalization of gender-responsive fiscal policies across budget cycle stages:
  1. setting of fiscal policy goals and targets;
  2. preparation of the annual budget and approval by the legislature;
  3. control and execution of the approved budget, and the collection of revenues;
  4. preparation of accounts and financial reports;
  5. independent oversight and audit of the budget.
- Gender-responsive budgeting does not require a new budgeting approach but an explicit recognition of gender elements paired with adaptation and reinforcement of existing institutions and tools.
- Key point: define critical entry points in the cycle for gender-related issues, include tools to assess policies, and assure approved policies are funded and implemented efficiently and effectively.

- Country examples of PFM initiatives:
  - Morocco: the “Gender Report” is central to gender budgeting, identifying key gender-gap areas, offering assessments, and setting goals. Early versions (2005-2007) included gender-based policy analysis; 2008 version examined performance indicators for operating and capital budgets; later reports added evaluation of measurable human rights indicators. Some ministries report sectoral- and sex-disaggregated data such as staff breakdown by sex or expenditures on women’s programs (Kolovich and Shibuya, 2016).
  - Uganda: introduced a Gender and Equity Compliance Certificate into their Public Finance Management Act in 2015 requiring ministries, departments, and agencies to ensure budget frameworks and Ministerial Policy statements address gender oriented goals; budgets are approved once a certificate reflects that gender goals were addressed (Stotsky et.al, 2016).

### Other policies (non-fiscal)
- Labor legislations:
  - Sri Lanka’s constitution supports non-discrimination in employment and occupation, but improving FLFP requires enhanced review and enforcement of labor law clauses, anti-harassment measures, and maternity leave—especially in the private sector.
  - Recommended activities (Solotaroff et.al, 2018; ILO, 2016):
    - Review labor laws for clauses that restrict women’s access to employment; introduce proper benefits and remuneration for flexible work (part-time, telework, work from home, night work) via legislative reform (amend Shop and Office Employees Act No. 15 of 1954 or issue new act).
    - Ensure laws are correctly interpreted for part-time and flexible work arrangements with benefits.
    - Expand application of maternity leave legislation to improve benefits and enforcement in the private sector and introduce paternity leave.
    - Enhance safety regulation and labor monitoring audits of the workplace.

- Access to financial services:
  - In urban areas access to finance is generally not an issue; but women in plantation sector (predominantly in North-east provinces) and other “pockets of poverty” need greater presence of financial institutions.
  - Strengthening women’s access to finance promotes entrepreneurship and labor market participation.
  - Recommendation: ensure access includes formal financial institutions in addition to microfinance.
  - Evidence: several studies show positive impact of female CEOs on firm profitability, management, and innovation (Christiansen et.al., 2016).

### D. Conclusions
- Increasing female labor force participation in Sri Lanka is macro-critical and urgent given demographic transition, rapidly ageing population and shrinking labor force.
- Drivers of low FLFP include:
  - low proportion of high skilled female workers and persistent skill mismatches,
  - outdated clauses in labor laws limiting part-time and flexible work,
  - absence of good quality and affordable child care facilities,
  - poor enforcement of laws providing social protection, especially in transportation.
- The Sri Lanka Government has formulated strategies to address these constraints.
- Additional measures include:
  - systematic engagement in gender responsive budgeting, leveraging international experiences;
  - non-fiscal measures to update labor law clauses and promote greater financial inclusion.

### Annex II. Measures Outlined in 2018 Budget to Achieve Gender Equality (selected items)
- Number of female entrepreneurs engaged in sustainable enterprise utilizing small and medium-sized enterprises (SMEs) loans/subsidies.
- Increased percentage of women participation in labor force.
- Increased number of female headed households that have built houses through financial assistance programs.
- Increased percentage of females who obtain National Vocational Qualifications (NVQ).
- Number of national policies and projects aimed at employing the female migrant returnees and potential migrant females.
- Increased number of females in decision making positions attained through policy interventions.
- Number of public institutions that have maintained gender disaggregation data system.
- Number of institutions that have taken measures to improve gender friendly work environment and day care facilities.
- Number of policies and guidelines are in place and effectively implemented to ensure gender concerns are addressed in rescue, relief, rehabilitation, and reconstruction stages of disasters.
- Number of police women and children bureau units of police stations are enabled to handle women and children issues in a sensitive and effective manner.
- Number of gender discriminatory laws, policies, and procedures that are amended, enacted and/or implemented.
- Amount of allocation made to complement Multi-Sectoral National Action Plan to address sexual and gender-based violence (SGBV).

*Source: cr18176 - 16.      In this context, some of the policy instruments can be considered as follows:*

### introduction of an automatic pricing mechanism for electricity, combined with well-targeted safety

### introduction of an automatic pricing mechanism for electricity, combined with well-targeted safety nets, can contain fiscal risks from droughts

### A. Introduction
- Sri Lanka is prone to natural disasters, with weather-related events (from severe floods to extreme droughts) being recurrent and more frequent than in other countries.
- Recent events:
  - Large-scale flooding in May 2016 followed by a historically severe drought in late 2016.
  - Monsoon in May 2017 triggered floods and landslides in the southwest, claiming more than 200 lives and damaging more than 9,000 houses.
- Impacts on 2016–17:
  - Agriculture GDP contracted by 3.8 percent and 0.8 percent in 2016 and 2017, respectively.
  - Food inflation accelerated to 14.4 percent y/y in December 2017.
  - Food imports increased by about US$200 million in 2017 (0.2 percent of annual GDP).
  - Drought shifted power generation from hydro to thermal, raising oil imports by about US$900 million (1 percent of GDP) in 2017.
  - More than 2 million people affected by the floods and drought.
  - Food-insecure households estimated at 277,000 in August 2017, with 30 percent of them possibly consuming less than the daily minimum calorie intake.
- Policy intent: Vision 2025 prioritizes environmental protection and disaster management, including a national disaster reserve fund for post-disaster reconstruction.
- Note structure: Section B reviews principles; Section C discusses contingency budgeting, electricity pricing reforms, and disaster-linked insurance; Section D concludes.

### B. Developing National Disaster Risk Management Framework
- Key elements of a comprehensive, multi-pillar risk management framework (IMF, 2016):
  - Identifying and assessing natural disaster risks.
  - Developing self-insurance through fiscal and external buffers.
  - Risk reduction via structural reforms and targeted infrastructure investments.
  - Risk transfer through disaster risk insurance, multilateral risk pools and precautionary instruments.
- Three-tiered financing approach (World Bank, 2012), by return period:
  - Low-risk layer (return periods of about 5 years or less): annual budget allocation or contingency budget; disaster-linked social protection for vulnerable households.
  - Medium-risk layer (return periods of about 5–20 years): contingent credit for quick post-disaster financing.
  - High-risk layer (return periods greater than 20 years): transfer to international capital and insurance markets (catastrophe bonds, catastrophe derivatives).
- Sri Lanka risk profile (World Bank, 2016; based on 1998–2012 historical data, excluding 2004 tsunami):
  - Combined average annual loss from natural disasters estimated at about 0.5 percent of GDP:
    - 0.32 percent of GDP for floods.
    - 0.05 percent of GDP for droughts.
    - Remainder for landslides and cyclones.
  - High-risk layer events: estimated disaster costs of about 2.4 percent of GDP once every 100 years.
- Strategic recommendation:
  - Develop a comprehensive disaster risk financing strategy consistent with public debt sustainability.
  - Near term: contingency budgeting and well-designed disaster insurance; an automatic electricity pricing mechanism to contain drought-related contingent liabilities.
  - Medium/long term: use contingent credit (e.g., informed by Cat-DDO experience), and develop parametric disaster insurance, catastrophe risk pools, and catastrophe bonds, requiring regulatory, implementation capacity, and infrastructure build-up.

Box — Innovative Transfer Options (summarized descriptions preserved)
- Parametric disaster insurance:
  - Pays out when third-party data confirm a disaster event based on pre-defined parameters rather than assessed incurred costs.
  - Advantages: fast post-disaster payouts; tailored contracts.
  - Challenges: data intensive; requires detailed asset databases and loss models; market still developing and cost can be high.
- Catastrophe risk pools:
  - Pool risks across regions or within a country to generate economies of scale and reduce operating, capital, and information costs.
  - Examples: Turkey’s Catastrophe Insurance Pool; sovereign catastrophe risk pools for the Caribbean and Latin America, Africa, and the Pacific.
  - Benefits: price risk, increase value of risk information, create incentives for risk reduction.
- Catastrophe (CAT) bonds:
  - Offer institutional investors high coupons; in the event of a disaster the principal is forgiven to free resources for disaster response.
  - Principal forgiveness depends on parametric triggers based on scientifically measurable hazard characteristics.
  - Examples: Mexico issued CAT bonds in 2006, 2009, 2012 via World Bank’s MultiCat Program; World Bank issued a CAT bond for the CCRIF.

### C. Near-Term Options to Mitigate Disaster Financing Risks

Contingency Budgeting
- 2017 central government disaster-related spending (actual) and 2018 (budgeted) categories and totals:
  - 2017 Actual: Total Rs 53.5 bn (0.4 percent of GDP)
    - Mitigation: Rs 0.4 bn (0.00 percent of GDP)
    - Rehabilitation: Rs 34.4 bn (0.26 percent of GDP)
    - Cash support: Rs 5.2 bn (0.04 percent of GDP)
    - Insurance: Rs 5.5 bn (0.04 percent of GDP)
    - Other: Rs 8.0 bn (0.06 percent of GDP)
  - 2018 Budget: Total Rs 21.0 bn (0.15 percent of GDP)
    - Mitigation: Rs 15.2 bn (0.11 percent of GDP)
    - Rehabilitation: Rs 0.6 bn (0.00 percent of GDP)
    - Cash support: Rs 0.0 bn (0.00 percent of GDP)
    - Insurance: Rs 3.5 bn (0.02 percent of GDP)
    - Other: Rs 1.7 bn (0.01 percent of GDP)
- Observations:
  - In 2017, rehabilitation comprised about two thirds of total disaster spending.
  - Rehabilitation and cash support were not appropriated in the original 2017 budget, necessitating a mid-year budget realignment.
  - The 2018 budget includes mitigation projects but omits appropriations for rehabilitation and cash support, implying potential future need for budget realignment and delayed disaster response.
- Contingency budgeting recommendations:
  - Incorporate a contingency budget for emergency cash support and infrastructure rehabilitation within the central government budget.
  - Size contingency budgets within the total spending envelope and consistent with fiscal balance targets to safeguard public debt sustainability.
  - Unused contingency funds can build reserves for future disasters, subject to stringent governance and transparency.
  - Medium-term development into a natural disaster reserve fund (e.g., Mexico’s FONDEN) is an option.

Box — Use of Dedicated Contingency Funds (key characteristics)
- Dedicated funds typically have:
  - Dedicated financing source, specific governance and investment rules, and restrictive utilization rules.
  - Consolidation with budget information or at minimum disclosure in financial statements and budget execution reports.
  - Standing appropriations triggered by events (e.g., declaration of disaster emergency).
  - Clear use rules, audited financial statements, defined governance, and prudent transparent investment policies; normal PFM rules should apply with adjusted procurement rules for rapid response.
  - Drawdowns limited to disasters with large fiscal impacts (thresholds or minimum total cost estimates).
  - Fund size determined by expected damages, likely international support, ability to borrow in an emergency, and opportunity costs; fund should not be too large.
  - Fund as a funding source, not an implementing agency; spending authority should rest with implementing agencies.

Box — Mexico’s Natural Disaster Fund (FONDEN) (summary)
- FONDEN created in 1996 to address post-disaster financing delays.
- Mandate:
  - Finance post-disaster emergency assistance (revolving fund).
  - Provide federal agencies and Mexican states with financial resources when losses exceed budget capacity.
- FONDEN finances recovery and reconstruction of public assets (100 percent federal assets; 50 percent state/municipal assets) and low-income houses.
- In 1999, FONDEN Trust Fund established to finance FONDEN via a catastrophe reserve fund that accumulates unspent disaster budgets.

Drought and Electricity Costs: Fiscal Risk Mitigation
- Structural facts and impacts:
  - Hydropower generation accounts for 30 percent of total electricity generation capacity in Sri Lanka.
  - Droughts reduce hydro utilization and increase thermal (coal and oil) generation needs, raising electricity supply costs.
  - The 2017 drought raised the average electricity cost to about Rs 20 per kWh.
  - End-user tariffs have not been adjusted since 2014; CEB incurred losses of 0.4 percent of GDP in 2017 due to elevated electricity costs.
- Policy recommendation:
  - Cost-reflective tariff adjustments combined with well-targeted cash transfers can mitigate fiscal risks.
  - Electricity subsidies tend to benefit the rich more than the poor and should be contained.
  - An automatic tariff adjustment mechanism under the 2015 tariff methodology (cabinet approval expected by September 2018) would minimize subsidy accumulation.
    - The 2015 tariff methodology fixes tariffs for six months to introduce price stability and reflects forecasts of supply and cost for the forthcoming six months to eliminate discrepancies between cost and price over time.
  - To protect poorer segments from distributional impacts of higher tariffs, use lifeline tariffs and/or targeted cash transfers.
  - Ongoing reforms to social safety nets to improve targeting and coverage are critical to support tariff reforms.

### C. Improving Natural Disaster Risk Insurance (brief)
- Well-designed disaster insurance and innovative risk transfer (parametric insurance, catastrophe risk pools, CAT bonds) can improve post-disaster financing; however, implementation requires data, regulatory capacity, and infrastructure.
- Contingent credit options (e.g., World Bank Cat-DDO experience) should be sized and triggered to safeguard public debt sustainability; note: Cat-DDO became effective in August 2014 with total amount of $102 million (0.13 percent of the 2014 GDP), and the full amount was drawn in August 2016 to finance May 2016 flooding.

*Source: cr18176 - introduction of an automatic pricing mechanism for electricity, combined with well-targeted safety nets, can contain fiscal risks from droughts*

### 13.      Public insurance schemes are currently in place to cover the population against

### cr18176 - 13.      Public insurance schemes are currently in place to cover the population against

### Public insurance schemes and scope of coverage
- The National Natural Disaster Insurance Scheme (NNDIS) was introduced in April 2016 and is administered by the state-owned National Insurance Trust Fund (NITF).
- Coverage:
  - All households and small- and medium-sized enterprises for property losses up to Rs 2.5 million caused by disasters including cyclones, floods, and earthquakes (droughts are excluded).
  - Premiums for the NNDIS are paid by the government.
- Risk-transfer arrangement:
  - NITF purchases reinsurance from the international reinsurance market to provide financial protection to their capital.
- Crop insurance:
  - Crop insurance schemes administered by state-owned insurers cover farmers against crop losses due to natural disasters including droughts and floods.

### Performance and fiscal impact of the NNDIS
- Losses and costs:
  - Flooding in 2016 and 2017 resulted in net losses for the NNDIS, estimated at about Rs 2 billion for 2016-17 (0.02 percent of GDP).
  - Reinsurance costs were about Rs 800 million for 2017.
  - Premium paid by the government for 2017 was Rs 500 million.
- Diagnostic observation:
  - Relatively high reinsurance costs were not fully offset by government-paid premiums, contributing to net losses.

### Options to improve NNDIS risk management (policy recommendations)
- Thoroughly assess NNDIS risk exposure based on asset values and average payouts to households.
- Minimize reinsurance costs by:
  - Carefully choosing risk layers to be covered.
  - Adjusting deductibles.
  - Setting appropriate coverage limits.
- Transition to a targeted scheme to focus support on poor and vulnerable households, for example:
  - Reducing the maximum property losses covered for richer households.
  - Encouraging wealthier households to purchase insurance from private insurers.
- Use the new Social Registry (being developed by the Welfare Benefits Board) to:
  - Collect data on assets to be insured.
  - Target poor households in a manner consistent with other government welfare programs.
- Improve institutional accounting and transparency:
  - NITF should clearly demarcate risk exposure by insurance schemes and functions (e.g., separate accounts between the NNDIS and the reinsurance).
- Technical assistance:
  - The World Bank is providing technical assistance in this area.

### Strengthening disaster insurance more broadly (policy recommendations)
- Consider introducing a catastrophe insurance program for public assets to facilitate post-disaster reconstruction of damaged infrastructure such as roads and bridges.
- Standardize insurance cover across key public assets to:
  - Generate economies of scale.
  - Achieve diversification benefits and thereby lower premiums.
- Improve crop insurance schemes by:
  - Assessing risk exposure.
  - Reviewing the payment trigger.
  - Evaluating potential benefits of risk pooling.
- 2018 Budget measure:
  - Announced establishment of a weather-indexed crop insurance scheme for rice paddies and other crops such as maize and soya, under a contributory scheme with the premium paid by both farmers and the government.

### Near-term and longer-term disaster risk financing recommendations (conclusions)
- Near-term measures (three highlighted):
  - Introduce a contingency budget for emergency cash support and infrastructure rehabilitation within the total spending envelope consistent with the fiscal balance target.
  - Introduce automatic pricing mechanism for electricity, combined with well-targeted safety nets, to contain fiscal risks from droughts.
  - Improve risk management of disaster insurance schemes to maximize effectiveness for post-disaster reconstruction while minimizing costs.
- Longer-term priorities:
  - Develop a comprehensive disaster risk financing strategy consistent with Sri Lanka’s debt sustainability.
  - Build up infrastructure and implementation capacity required for innovative risk transfer approaches such as parametric insurance, catastrophe risk pools, and catastrophic bonds.

*IMF staff summary of chapter content (paras 13–16).*

### 8.      Sri Lanka’s significant credit growth for more than two consecutive years warrants

### 8.      Sri Lanka’s significant credit growth for more than two consecutive years warrants attention.

### Credit growth, risks, and current assessment
- Significant credit growth for more than two consecutive years warrants attention.  
- Part of the credit growth can be associated with financial deepening, but loan concentration, notably in the real estate sector, raises concerns.  
- Current macro-financial indicators do not point to systemic risks but low NPLs may not be reflective of the underlying vulnerabilities, calling for heightened vigilance.  
- Authorities’ ongoing efforts that will strengthen the financial sector and help maintain financial stability: adoption of new banking supervision and regulation practices (Basel III); improvements to financial infrastructure; revisions to the legal framework relating to the financial system (such as the Banking Law Act).  

### Short-term policy recommendations (financial stability / monetary stance)
- The Central Bank of Sri Lanka (CBSL) should maintain a prudent data-dependent monetary stance in the short-term to ensure that excessive credit growth does not lead to inflationary pressures.  
- Adopt an overall macro-prudential framework to strengthen financial stability. Possible elements include:  
  - loan-to-value ratio limits;  
  - higher risk weights on mortgages and construction;  
  - calibration of measures to the specific needs of Sri Lanka.

### Medium-term policy recommendations (supervision, regulation, resolution)
- Continue policy efforts to strengthen:  
  - supervision and regulation of both banking and non-bank financial institutions; and  
  - resolution frameworks for both banking and non-bank financial institutions.

### References and analytical foundations (selected)
- Basel Committee on Banking Supervision (2010). Guidance for national authorities operating the countercyclical capital buffer, December.  
- Borio, C. and M. Drehmann (2009). Assessing the risk of banking crises – revisited, BIS Quarterly Review, March, pp 29–46.  
- Drehmann, M., C. Borio and K. Tsatsaronis (2011). Anchoring countercyclical capital buffers: the role of credit aggregates, Bank for International Settlements Working Paper No. 355.  
- IMF (2011). Global Financial Stability Report: Grappling with Crisis Legacies, Chapter 3: Toward Operationalizing Macroprudential Policies: When to Act?  
- Jones M. T. and M. Karasulu (2006). The Korean Crisis: What Did We Know and When Did We Know It? What Stress Tests of the Corporate Sector Reveal, IMF Working Paper 06/114.  
- Lindner P. and S. E. Jung (2014). Corporate Vulnerabilities in India and Banks’ Loan Performance, IMF Working Paper 14/232.

---

### Fiscal rules background and performance
- The Fiscal Management (Responsibility) Act (FM(R)A) was legislated in 2003 and subsequently revised in 2013 and 2016. The FM(R)A contains rules on debt, budget deficit and government guarantees.  
- The Act contains few enforcement mechanisms and does not prescribe corrective action in the event any of the targets are breached—it only requires that the government explain to Parliament the reason for the departure, steps the government plans to take to “overcome the causes necessitating such departure,” and the amount of the time the departure will last. Although the Act allows for departure only under “exceptional circumstances” and with the approval of Parliament, no formal escape clauses are specified and in practice the thresholds are serially breached with little consequence.  
- Summary of key provisions in the FM(R)A (original and amendments):  
  - Original (2003): Budget Deficit — Not greater than 5% of GDP by end 2006 and thereafter. Debt — Not greater than 85% by end 2006, less than 60% by end 2013. Guarantees — 3-year moving average not exceeding 4.5% of GDP.  
  - Amendment (2013): Budget Deficit — Unchanged. Debt — Not greater than 80% by end 2013, 60% by end 2020. Guarantees — Limit revised to 7.0% of GDP.  
  - Amendment (2016): Budget Deficit — Unchanged. Debt — Unchanged. Guarantees — Limit revised to 10% of GDP.

### Historical outcomes against the rules
- The rules have not been effective in achieving their objectives:  
  - Central government debt ratio fell from around 106 percent in 2002 to around 88 percent of GDP by 2006, but never met the original objective of the rule.  
  - Debt was around 72 percent of GDP in 2013 but nearly 80 percent of GDP by 2017.  
  - The overall deficit has never been less than 5 percent of GDP over the period from 1990 to 2017.  
- Early-years deviations were exacerbated by shocks: the December 2004 Tsunami, rising commodity prices and the global financial crisis in 2007–08, and increased military spending at the end of the civil war (with resettlement, reconstruction and rehabilitation costs in 2008-09). Deviations since 2013 are harder to attribute to macroeconomic shocks and reflect limited political consequences for breaking rules and perceptions that targets are unrealistic.

### Principles underpinning effective fiscal rules
- Key criteria for effective fiscal rules (IMF 2018a):  
  - Sustainability. Compliance with the rule should ensure long-term debt sustainability.  
  - Stabilization. The rule should let automatic stabilizers operate and/or allow discretionary countercyclical changes in taxes or expenditures.  
  - Simplicity. The rule should be easily understood by decision makers and the public.  
  - Operational Guidance. The rule should translate into clear guidance in the annual budget process; budget aggregates targeted should be largely under the control of the policymaker.  
  - Resilience. A rule should be in place for a sustained period to build credibility and should not be easily abandoned after a shock.  
  - Ease of monitoring and enforcement. Compliance should be easy to verify and there should be costs associated with deviations.

### Choice of operational rules and suitability for Sri Lanka
- Best practice: fiscal frameworks built around two pillars — a fiscal anchor (e.g., debt-to-GDP) and an operational rule on fiscal aggregates.  
- Golden rules and adjusted budget balance rules are likely less well suited to Sri Lanka:  
  - Golden rule protects capital spending but is a weak speed limit on debt accumulation.  
  - Cyclically adjusted/structural rules require estimating the output gap which is inherently uncertain in a small open economy like Sri Lanka.  
- Expenditure rules: improve fiscal stability and are easy to communicate but have weaker links to debt sustainability and can shift adjustment onto infrastructure and social spending.  
- Revenue rules: can help raise revenue or limit tax burden but may be procyclical.

### Current operational rule assessment
- Sri Lanka’s current fiscal rule appropriately uses the overall budget deficit as the operational target. Advantages: simplicity and direct link to debt dynamics. Disadvantages: potential pro-cyclicality and incentives to cut capital spending; primary balance variant excludes interest payments but can weaken link to debt sustainability if thresholds are not regularly reassessed.

### Rule calibration methodology and outcomes
- Calibration principles (IMF’s How-to Note, IMF 2018b): comprehensive and consistent; sequenced (set debt ceiling first, then operational rule); prudent (account for fiscal risks and preserve buffers); updated regularly but not too frequently.  

Step 1 — Setting a debt limit:
- For the simulations a debt limit of 70 percent of GDP over the medium term is assumed. This corresponds to the Vision 2025 target and the debt distress threshold for emerging market economies in the IMF DSA methodology.

Step 2 — Estimating effects of shocks (stochastic simulations):
- Simulations use a joint distribution of macroeconomic variables (GDP growth, interest rates on government debt (foreign and domestic), exchange rate, contingent liabilities). The initial simulation year is 2021 and projections up to 2020 follow the May 2018 DSA.

Assumptions used for simulations (values in 2020):
- Debt-to-GDP 79.5%  
- Overall budget balance (% GDP)  -3.5%  
- Primary budget balance (% GDP) 2.1%  
- Real GDP growth 4.7%  
- Real effective interest rate (domestic) 6.6%  
- Real effective interest rate (foreign) -0.2%  
- Foreign debt share  44%

- Each simulation produces a medium-term path for macro variables subject to shocks; debt trajectories are derived from the debt accumulation equation and a fiscal reaction function where the primary balance may respond to debt and macro realizations. Fiscal shocks are included each period.  
- Stochastic-simulation results summarized in a fan chart indicate that a debt anchor of 50 percent is sufficient to ensure that debt stays below the debt limit of 70 percent with 90 percent certainty.

Step 3 — Calibrating the operational rule:
- Debt and deficits are tied through the debt dynamics accounting identity and choices of overall deficit limits were analyzed to assess implied years of convergence to the debt target. The initial year is set at 2021 with an initial debt level of 79.5 percent of GDP. The baseline scenario assumes an unchanged policy with an overall deficit of 3.5 percent of GDP from 2021 onwards.  
- Under the baseline scenario, Sri Lanka’s public debt will reach the debt anchor (50 percent of GDP) only by 2034. Sensitivity analysis: with higher nominal growth (12 percent versus 10 percent in the baseline) convergence could happen 5 years sooner. With slower growth or larger operating deficits, convergence would be beyond 2050.

*International Monetary Fund — cr18176 (section 8 and subsequent fiscal-rules analysis).*

### 18.      Pivoting to a smaller overall deficit after the initial ceiling is reached would accelerate

### 18.      Pivoting to a smaller overall deficit after the initial ceiling is reached would accelerate

### Convergence scenarios and key quantitative findings
- Under an active policy scenario that gradually transitions from a 3.5 percent overall deficit in 2020 to 2 percent, debt would reach 50 percent of GDP by 2029 – five years faster than under the baseline.
- Baseline statement: achieving the fiscal consolidation path outlined in the May 2018 DSA up to 2020 and then gradually reducing the overall deficit from 3.5 percent of GDP to 2 percent over the period from 2021 to 2026 would accelerate convergence to the debt anchor by 2029 and reduce risks to debt sustainability.
- Note in the calibrations table: The smaller deficit scenario assumes gradual reduction in deficits from 3.5 percent in 2021 to 2 percent in 2025.

- Convergence table (In percent of GDP, unless otherwise noted):
  - Assumptions / Convergence Scenario / Initial Debt Level / Overall Balance / Nominal GDP Growth Rate (%) / Year Debt Ratio < 70% / Year Debt Ratio < 60% / Year Debt Ratio < 50%
  - Baseline / 79.5 / -3.5 / 10.0 / 2023 / 2027 / 2034
  - Stronger growth / 79.5 / -3.5 / 12.0 / 2022 / 2025 / 2029
  - Smaller deficit / 79.5 / -2.0 / 10.0 / 2023 / 2025 / 2029
  - Weaker growth / 79.5 / -3.5 / 8.0 / 2025 / 2033 / 2052
  - Larger deficit / 79.5 / -5.0 / 10.0 / 2026 / 2037 / --

- Footnotes and technical caveats:
  - The initial debt level is the 2020 figure from the May 2018 DSA.
  - In practice the link between debt and deficits can be influenced by accumulation of financial assets, currency fluctuations, and non-debt financing of deficits but these effects tend to temporary.
  - Since the overall balance would be a target for the central government this is equivalent to the SOE sector running balanced budget throughout the project.

### Design features for effective fiscal rules (policy guidance)
- Core recommendation:
  - A second-generation fiscal rule for Sri Lanka should be anchored by the debt ratio and utilize the overall balance as an operational target, emphasizing simplicity and operational guidance.
- Calibration guidance:
  - Latest calibration techniques suggest a debt anchor of 50 percent of GDP would be appropriate to ensure debt stays below 70 percent of GDP with 90 percent probability.
- Rule design principles to adopt:
  - Broad institutional and economic coverage.
  - Escape clauses to deal with unforeseen events.
  - Automatic correction mechanisms to define what should be done in the case of a breach.
  - Effective monitoring often carried out by an independent fiscal council.

### Enhancing compliance: corrective mechanisms and triggers
- Corrective mechanisms should have two parts: the trigger and the correction action.
- Trigger design considerations:
  - Triggers can activate after a breach or when there is an elevated risk of breach.
  - Breach assessment can be quantitative (threshold crossed) or qualitative (fiscal council or policymakers determine a significant breach).
  - The assessment period matters: triggers can respond to single-year deviations, averages over two years (Finland, Ireland and Italy), or cumulative deviations (Germany).
- Correction action design considerations:
  - Size of the corrective action: from forcing the operational target back below threshold to fully compensating the deviation (undershooting by the amount of the breach in subsequent years).
  - Timeframe for correction: immediate (in-year or next year's budget) to several years; shorter horizons near debt limits.
  - Policy mix: full discretion for policy makers (revenues or expenditure) or prescribed specific actions (e.g., across-the-board expenditure cuts).

### Coverage, escape clauses, and monitoring
- Coverage:
  - Fiscal rules should be broad-based to prevent circumvention via fiscal or quasi-fiscal activities (e.g., SOE activities).
  - A broad rule could cover the nonfinancial public sector, including state enterprises (example cited: Brazil).
- Escape clauses:
  - Well-designed escape clauses should:
    - Have a limited and clearly defined set of triggering events (and should not include cyclical events).
    - Include time limits on deviations from targets under the clause.
    - Require fiscal policy to return to targets after the escape clause operation ends.
- Monitoring and institutions:
  - Strong fiscal forecasting capabilities and credible monitoring institutions underpin effective rules.
  - Dedicated macro-fiscal units improve forecasting and monitoring.
  - Independent fiscal councils can increase political incentives to adhere to rules by raising the cost of non-compliance; they must be truly independent and credible.
  - Examples of countries using fiscal councils outside the EU: Brazil, Chile, Colombia, and Peru.

### Conclusions and policy recommendations (summarized)
- A debt-anchored rule with the overall balance as operational target is appropriate for Sri Lanka, emphasizing simplicity and operational guidance.
- Calibration suggests a 50 percent of GDP debt anchor to keep debt below 70 percent with 90 percent probability.
- Achieving the May 2018 DSA consolidation to 2020 and then reducing the overall deficit from 3.5 percent to 2 percent would accelerate convergence to the debt anchor by 2029 and reduce debt sustainability risks.
- The second-generation fiscal rule should include:
  - Appropriately designed enforcement mechanisms (correction mechanisms) and escape clauses.
  - Consideration of strengthening macro-fiscal forecasting through a dedicated unit.
  - Utilizing a fiscal council to independently monitor compliance.

### Annex I — Direct calibration method for the debt ceiling (summary)
- Approach:
  - Characterize joint distribution of macro variables (growth, average interest rate on debt, exchange rate) by calibrating a multivariate normal (or student-t) distribution from historical co-movements.
  - Draw N sequences of 6-year projections by repeated sampling from the calibrated multivariate distribution.
- Steps to calibrate debt ceiling:
  - Draw macroeconomic variable forecasts N times to create N projection sets.
  - Generate N trajectories of the primary balance using a fiscal reaction function and previous-year debt; fiscal shocks distribution calibrated from deviations between actual and predicted primary balances.
  - Obtain N debt trajectories from the debt accumulation equation (government budget constraint) including a constant stock-flow adjustment (SFA).
  - Adjust the “starting level” of debt by small amounts (0.4 percent or below) and repeat until the 95th percentile of the debt distribution falls into a small interval around the maximum debt limit (MDL): ݐܾ݁ܦଽହ ∈ (−0.4ܮܦܯ; 0.4ܮܦܯ). The starting level satisfying this is the debt ceiling; safety margin = MDL − debt ceiling.

- Debt accumulation equation as presented:
  - ݀௧ = [1 + (ݎ௧ − ݃௧) + ... ] (presentation in source; debt ratio evolves with average effective real interest rate, real GDP growth rate, primary balance, and stock-flow adjustment).

### Annex II — Fiscal Reaction Function (FRF) (summary)
- FRF specification (Bohn (1998) style) links primary balance to fiscal and macro conditions.
- Estimated FRF form (as presented):
  - ܾ݌௜௧ ߙ = ܾ݌௜௧ିଵ ߚଵ + ... (full equation reproduced in source).
- FRF used in simulations estimated across 25 EMs.
- Coefficient estimates as provided:
  - ߙൌ	െ0.027,  ߚଵൌ0.466,  ߚଶൌ0,  ߚଷീ0.199,  ߩൌ0.0645.

### Annex III — Stock-Flow Adjustment (SFA) estimation (summary)
- SFA estimated as the difference between fitted values of the debt equation and actual debt data (ratios to GDP).
- Key variables in SFA expression:
  - d = debt-to-GDP ratio; g = real growth rate; dα and fα = ratios of domestic and foreign currency denominated to total debt; dtr and ftr = real interest rates on domestic and foreign currency loans; εΔ = real exchange rate depreciation.
- SFA can account for constant realizations of contingent liabilities each period.

_International Monetary Fund — Sri Lanka: Convergence Scenarios for the Debt Ratio (excerpt)_

---


_Source: https://www.imf.org/-/media/files/publications/cr/2018/cr18176.pdf_