Sri Lanka Spent Only 8% of Cyclone Ditwah Rebuilding Funds as Budget Turned to Surplus

Rs. 460 billion remained unspent as reconstruction lagged

Sri Lanka had disbursed only about Rs. 40 billion of its Rs. 500 billion allocation for reconstruction after Cyclone Ditwah by the middle of 2026, leaving nearly Rs. 460 billion unspent. The slow release of money helped produce a rare budget surplus even as damaged communities awaited rebuilding.

The World Bank's October 2026 Sri Lanka Development Update, titled From Recovery to Transformation, puts the reconstruction delay within a wider failure to turn budget allocations into spending. Only 16.7 percent of the annual capital budget had been used by June, while less than half of a separate Rs. 100 billion relief package linked to the Middle East war had been paid out.

The Bank attributes the shortfall to weaknesses in project selection, procurement and implementation capacity, rather than a shortage of financing. That distinction matters: approving money does not ensure that projects are ready, contracts can be awarded or work can proceed.

At the same time, revenue rose 27.2 percent from a year earlier, far outpacing the 7.9 percent increase in expenditure. The government recorded a budget surplus of Rs. 9.5 billion in the first half of 2026, compared with a Rs. 406 billion deficit in the same period of 2025.

The figures reveal two different sides of Sri Lanka's recovery. Tax collection and fiscal balances have strengthened, but the institutions responsible for delivering reconstruction and public investment have struggled to keep pace. A better Treasury balance therefore does not, by itself, establish that the recovery is reaching damaged homes, businesses and infrastructure.

The damage behind the reconstruction budget

Cyclone Ditwah struck between late November and early December 2025, bringing heavy rain, flooding and landslides across all 25 districts. Central and south central areas, including Kandy, Badulla, Nuwara Eliya, Kurunegala and Matale, suffered severe damage and difficulties reaching affected communities.

In its January 9, 2026 reconstruction policy analysis, ODI cited United Nations figures showing that nearly 10 percent of the population had been affected and more than 91,000 homes damaged or destroyed. Its account put the number of people dead or missing at more than 600 at that stage.

The World Bank's Global Rapid Post-Disaster Damage Estimation assessed damage at US$4.1 billion, approximately 4 percent of Sri Lanka's 2024 gross domestic product. Infrastructure represented 42 percent of damage, residential buildings 24 percent, agriculture 20 percent and other buildings 14 percent.

Those shares show why compensation payments alone cannot describe the reconstruction effort. Roads, railways, energy systems and agricultural assets also require repair. Each category involves different agencies, project preparation and contracting arrangements.

The strain had already become evident during the cyclone and war shocks described in April 2026. At that point, the disaster was colliding with higher fuel costs and the continuing effects of Sri Lanka's 2022 economic crisis. The October spending figures show that the subsequent challenge was also one of converting allocated funds into assistance and completed work.

Household compensation follows a different measure

A separate account of compensation provides a more advanced spending figure, but covers a narrower allocation. Prageeth Danansuriya, Senior Assistant Secretary of the National Disaster Relief Services Centre, said Rs. 150 billion had been allocated through the budget to compensate for damage to houses, business premises and other buildings.

Danansuriya said Rs. 39 billion had been spent, describing that as around 25 percent of the allocation. Using the stated amounts, Rs. 39 billion represents 26 percent of Rs. 150 billion, consistent with an approximate description.

This is not directly comparable with the World Bank's finding that about 8 percent of the Rs. 500 billion reconstruction allocation had been disbursed by June. The compensation figure concerns a specific category, and its reporting cutoff is not established. It cannot be treated as an updated spending rate for the entire reconstruction programme.

Neither figure establishes how many homes had been repaired or rebuilt. Money paid out, compensation received and construction completed are different measures. The published accounts do not provide a common breakdown that would allow readers to track all three.

Spending remained below 30 percent through September

The delay extended beyond the first half of the year. Figures attributed to two government sources put capital spending at around Rs. 385 billion through September, against a 2026 allocation of Rs. 1,380 billion. That is approximately 27.9 percent of the annual budget after nine months.

On those figures, roughly Rs. 995 billion remained to be spent in the final three months if the government was to use the full allocation. That would require average monthly expenditure of about Rs. 332 billion, compared with approximately Rs. 43 billion a month during the first nine months.

The comparison illustrates the scale of the required acceleration, rather than proving that spending should be equal every month. Payments can be concentrated late in the year as contracts progress and invoices fall due. Nevertheless, reaching the full allocation would require a sharp change in the pace recorded through September.

The two government sources said the administration expected to use the full capital allocation before the end of 2026. Cabinet Spokesman and Media Minister Nalinda Jayatissa said on October 6 that 70 percent of spending is finalized in November, while saying the government could not provide an updated total that day.

The main dates in the spending and recovery record are:

• Late November to early December 2025: Cyclone Ditwah caused widespread flooding, landslides and damage.
• End of December 2025: Emergency foreign aid commitments totaled approximately US$818 million.
• June 2026: About 8 percent of reconstruction funds and 16.7 percent of the annual capital budget had been spent.
• September 2026: Capital expenditure reportedly reached approximately Rs. 385 billion, or 27.9 percent of the annual allocation.
• November and December 2026: The government expects spending to accelerate and the full capital allocation to be used by year end.

Stronger taxes helped create the surplus

The revenue improvement was substantial. Tax collections rose 25.9 percent in the first half of 2026, mainly driven by Value Added Tax, or VAT, and duties on vehicle imports after import restrictions were lifted. VAT is a tax collected on sales of goods and services, while import duties are charged on goods entering the country.

The primary surplus reached Rs. 1,233.1 billion, about Rs. 360 billion higher than a year earlier. A primary balance excludes interest payments, making it useful for showing whether revenue covers the government's other spending. The much smaller budget surplus of Rs. 9.5 billion includes interest costs.

The difference between the two balances shows how heavily debt servicing still weighs on public finances. It also explains why a large primary surplus can coexist with only a narrow surplus once interest is counted.

Tax laws certified in June raised capital gains taxes, introduced VAT on foreign digital service providers and increased VAT on financial services. The government is also piloting a national electronic invoicing system. The accounts do not quantify how much each June measure contributed to revenue during the first half.

A legal ceiling limits how fiscal space can be used

Lower interest costs do not give the government unrestricted room to raise spending. The Public Financial Management Act caps primary expenditure, meaning spending excluding interest, at 13 percent of GDP.

Interest payments absorbed around 46 percent of government revenue in 2025, down from nearly 80 percent at the crisis peak between 2022 and 2023. The World Bank projects that share will fall to about 34 percent by 2028, a further reduction of 12 percentage points from 2025.

The Bank argues that the expenditure ceiling makes the effectiveness of each rupee more important. In the development update, it states:

This spending ceiling increases the importance of improving spending quality rather than simply increasing spending levels,

The legal limit and the implementation shortfall are separate constraints. The ceiling governs how much primary spending is permitted. Weak preparation, procurement and delivery can prevent the government from using even the amounts it has already budgeted.

The World Bank expects primary surpluses of at least 3.5 percent of GDP over the medium term. That is at least 0.9 percentage points above the government's 2.6 percent target from 2027 in its Fiscal Strategy Statement. The Bank also says some fiscal space could be used to reduce regressivity in the tax mix, meaning a tax burden that falls more heavily on poorer people relative to their income.

The proposed fixes focus on project delivery

The World Bank calls for a more systematic process of project screening, risk filtering, prioritisation and appraisal to reduce delays and cost overruns. In practical terms, that means testing whether a project is necessary, feasible, affordable and ready before committing money to it.

This recommendation connects the low spending figures with the government's growth ambitions. An allocation produces little immediate benefit if a project lacks a workable design, clear responsibilities or the capacity to manage a contract. Accelerating payments without addressing those weaknesses would not necessarily produce better infrastructure.

Ganeshan Wignaraja, who wrote ODI's January analysis, proposed a costed reconstruction strategy for the Rebuild Sri Lanka Fund, prepared with the World Bank and Asian Development Bank. His recommendations included stronger early warning systems, infrastructure able to withstand disasters, insurance and better institutional capacity.

Wignaraja also proposed a dedicated disaster management agency or ministry and discussions with the International Monetary Fund about modest adjustments to fiscal targets. These were policy recommendations, not decisions confirmed in the October accounts. The immediate test remains whether the government can prepare and carry out the projects already funded.

Foreign commitments are not the same as reconstruction delivered

ODI's January assessment listed approximately US$818 million in emergency aid commitments at the end of December 2025, around 20 percent of the World Bank's US$4.1 billion damage estimate. The comparison showed the scale of damage relative to announced support, not the amount already received or spent.

The listed multilateral support included US$206 million approved by the IMF under its Rapid Financing Instrument, US$120 million committed by the World Bank through funds redirected from existing projects, and US$43 million from the Asian Development Bank for trade finance and disaster relief.

Bilateral commitments included US$450 million from India, US$2.5 million from Japan, US$2 million from the United States and US$143,000 from China. These forms of assistance served different purposes and cannot all be read as direct payments for construction.

The IMF had postponed its fifth programme review to allow an assessment of cyclone damage, according to the January analysis. No subsequent review date is given in these accounts. Nor is there a consolidated October figure showing how much of the foreign support had been disbursed.

The foreign commitments and the Rs. 500 billion domestic reconstruction allocation also use different currencies and cover different categories. Without a common accounting basis, they should not be added together as if they represented a single pool of completed reconstruction spending.

Economic recovery has not removed household hardship

The World Bank reports that real GDP returned to its level before the 2022 crisis, with the economy growing 4.7 percent in the first half of 2026 after 12 consecutive quarters of expansion. It forecasts growth of 4.4 percent for 2026 and 4.2 percent for 2027.

Those gains have not been evenly shared. The Bank estimates poverty at 16.9 percent in 2025, still above the level before the crisis, and says labour market indicators continue to lag. Renewed inflation has also reduced some household welfare gains.

ODI's January analysis used an income poverty estimate of around 22 percent. That differs from the World Bank's later 16.9 percent estimate. The accounts do not establish whether the difference reflects revised data, different poverty thresholds or another methodological distinction, so the figures should not be presented as a measured fall in poverty.

The development update identifies agribusiness as one potential route to stronger growth and employment. Primary agriculture accounts for 8 percent of GDP, but the wider agrifood system contributes about one sixth of output, more than 40 percent of employment and nearly one third of merchandise exports.

That economic reach makes cyclone damage to agriculture and infrastructure relevant well beyond farms. The Bank identifies infrastructure gaps, restricted access to land and finance, weak investment incentives and limited processing of agricultural products as constraints. Timely reconstruction can address part of that problem, but the spending figures do not yet show whether those repairs are being delivered.

What to Know

• Only about 8 percent of the Rs. 500 billion Cyclone Ditwah reconstruction allocation had been disbursed by June 2026.
• Nearly Rs. 460 billion remained unspent, with the World Bank blaming implementation weaknesses rather than financing shortages.
• Capital spending reportedly reached Rs. 385 billion through September, approximately 27.9 percent of the annual allocation.
• Revenue growth and slow spending helped produce a Rs. 9.5 billion budget surplus in the first half.
• A separate Rs. 39 billion compensation figure covers a narrower allocation and does not establish the pace of the entire rebuilding programme.
• The Public Financial Management Act caps primary expenditure at 13 percent of GDP.
• The government expects spending to accelerate before year end, but completed reconstruction and final expenditure totals are not yet established.