Sri Lanka Customs collected a record Rs.2.56 trillion in 2025 and is running ahead of its 2026 target. But behind the impressive numbers, Auditor General reports reveal billions in unpaid revenue, valuation problems, weak controls and cases where the State collected far less than its own calculations said it was owed.
By Lakbima News Investigative Desk
September 8, 2026
There is a moment at a Sri Lankan Customs checkpoint that most people never see.
A container arrives.
Inside it may be clothing, electronics, machinery, food, cigarettes, spare parts or something much more valuable.
On paper, the shipment has a price.
That price determines taxes.
And those taxes become Government revenue.
But what happens if the declared value is wrong?
What happens if an importer declares a product under the wrong category?
What happens if goods are released before a dispute over their classification is settled?
What happens if a penalty is reduced?
What happens if millions in Customs revenue are technically owed but remain uncollected for years?
And what happens when the Government is trying to raise every possible rupee after an economic crisis?
These questions are not theoretical.
Sri Lanka’s own Auditor General has repeatedly identified weaknesses inside the Customs system that can affect the collection of public revenue.
At the same time, Sri Lanka Customs is reporting record-breaking collections.
In 2025, Customs says it collected approximately Rs.2.56 trillion, its highest revenue figure ever.
By June 30, 2026, Customs had already collected Rs.1.379 trillion, against a target of Rs.1.061 trillion — achieving 130% of its target.
So which story is true?
Is Customs performing exceptionally well?
Or is Sri Lanka still losing billions through weaknesses that are difficult to see from the headline revenue number?
The uncomfortable answer may be:
Both can be true.
A record year — but not the whole story
The first thing that must be made clear is that Sri Lanka Customs is not currently reporting a collapse in revenue collection.
Quite the opposite.
Customs reported its highest-ever annual revenue in 2025 at approximately Rs.2,557.5 billion.
And the 2026 figures presented to Parliament are equally striking.
By the end of June:
Expected revenue: Rs.1,060.559 billion
Actual revenue: Rs.1,379.084 billion
That is approximately 130% of the target.
Vehicle imports are a major reason.
Officials told Parliament that 316,000 vehicles had been imported by June 30, generating approximately Rs.512.547 billion in tax revenue.
Motor vehicles alone generated approximately Rs.386.726 billion in Customs tax revenue.
Petrol cars below 1,000cc generated approximately Rs.137.4 billion, representing 9.96% of total Customs revenue.
These are impressive numbers.
But revenue collected is not necessarily the same thing as revenue that could have been collected.
That distinction is at the heart of this investigation.
What exactly is “revenue leakage”?
Imagine two importers bringing identical goods into Sri Lanka.
Importer A declares the correct value and classification.
Importer B declares the goods at a lower value or under a different classification.
If Customs accepts the second declaration, both containers leave the port.
But the Government receives different amounts of tax.
The difference is revenue leakage.
It does not always mean corruption.
It can arise from:
deliberate undervaluation;
incorrect classification;
mistakes;
outdated valuation information;
exemptions;
weak enforcement;
delayed investigations;
unpaid penalties;
poor coordination between agencies;
or deliberate fraud.
And that distinction matters.
A Customs officer making a mistake is not the same thing as an importer deliberately deceiving the Government.
A system failing to collect a debt is not automatically corruption.
But repeated weaknesses can create opportunities for people who are willing to exploit them.
Customs itself knows undervaluation is a problem
Sri Lanka Customs established the National Import Valuation Database Unit, or NIVDU, specifically to improve import valuation.
Customs says the unit’s objectives include improving risk assessment and curtailing State revenue loss.
Its stated aim is to develop a valuation database that gives Customs officers reference values so they can assess imported goods more effectively and reduce undervaluation.
That tells us something important.
Undervaluation is not an imaginary problem invented by critics.
Customs itself has built an institutional mechanism specifically designed to combat it.
The question is how effective that mechanism is.
The billions that were owed — but not collected
The Auditor General’s 2024 examination of Sri Lanka Customs provides one of the clearest windows into this question.
The audit found that outstanding import tax dues amounted to approximately Rs.22.04 billion.
That represented about 38% of total outstanding tax revenue.
But during the year, Customs collected only approximately Rs.1.5 million from that amount — around 0.007%.
Read that number again.
Rs.22 billion was outstanding.
The amount recovered during the year was tiny by comparison.
There was another category that was even more striking.
Special commodity taxes outstanding amounted to approximately Rs.15.24 billion, around 26% of total outstanding tax revenue.
The Auditor General reported that no amount had been recovered from this category during the previous four years.
There was also approximately Rs.9.94 billion in Port and Airport Development Tax arrears.
Almost the entire amount had reportedly been outstanding for more than four years.
Only about Rs.1.5 million had been recovered.
And overall?
Customs began the year with approximately Rs.58.48 billion in revenue arrears.
Only about Rs.771.45 million was recovered during the year.
That was a recovery rate of approximately 1.32%.
This is not necessarily money that disappeared.
It is money the State says was owed.
But from the perspective of a Government struggling to finance schools, hospitals, roads and public services, there is an obvious question:
What happens when the State is good at collecting new revenue but slow at collecting old revenue?
The Rs.57 billion problem
The audit found another category of arrears that raises an even more uncomfortable question.
As of December 31, 2024, approximately Rs.57.72 billion was reportedly owed to Customs by various Government ministries and departments.
According to the audit report, these arrears related to amounts dating from before 2020.
Customs had sought approval to waive the amount under the relevant financial regulations.
This is not private-sector smuggling.
It is not an importer hiding goods.
It is money owed by Government institutions themselves.
The Auditor General noted that clearing the amount would require the respective institutions to seek additional budget allocations, potentially increasing Government expenditure.
In other words:
One arm of the State owes money to another arm of the State.
And ultimately, both arms are funded by the public.
The Rs.7.4 billion vehicle question
One of the most significant findings in the 2024 audit concerns vehicles assembled in bonded warehouses.
The audit examined seven manufacturing companies operating bonded warehouses.
In 2024, these companies assembled 538 motor cars and 53,381 motorcycles.
According to the Auditor General’s calculation, the production tax that should have been collected was approximately:
Rs.10.744 billion.
But Customs had collected approximately:
Rs.3.357 billion.
The difference was approximately:
Rs.7.388 billion.
That is not a rounding error.
It is billions.
The audit specifically identified this as a loss of excise-duty revenue.
Customs agreed with the audit observation.
This does not prove that anyone stole Rs.7.388 billion.
It does not prove criminal intent.
But it does prove something much more basic:
The Auditor General calculated that the State collected substantially less tax than it should have collected in this area.
That alone deserves public scrutiny.
How does a Customs valuation become a problem?
The Customs system has to determine the value and classification of imported goods.
That sounds straightforward.
It isn’t.
A product may have:
a manufacturer’s price;
a wholesale price;
a discounted price;
freight costs;
insurance;
related-party pricing;
different specifications;
different quality levels;
different country-of-origin costs.
Customs officers therefore need reliable information.
That is why the National Import Valuation Database exists.
But the more complicated the product becomes, the more room there may be for disagreement.
And disagreement can mean delay.
Delay can mean goods sitting in ports.
Or goods being released against guarantees.
Or revenue remaining unpaid.
The container problem
Sri Lanka imports enormous quantities of goods.
The 2026 Parliamentary Committee on Ways and Means specifically examined the volume of imported containers, how containers are inspected and the introduction of modern technology to improve Customs operations.
That raises a practical problem.
Customs cannot physically open every container.
If every container were manually inspected, Colombo Port and other entry points could grind to a halt.
So Customs must use risk management.
Some containers receive greater scrutiny.
Others are cleared faster.
That is normal international Customs practice.
But the system creates a vulnerability.
If criminals can predict what triggers an inspection, they can potentially structure shipments to avoid attention.
This is why intelligence and technology matter.
Not every container needs to be opened.
But the right containers need to be identified.
Technology could be the difference
Parliament was told that Customs is looking at modern technology to improve efficiency.
That matters because the modern Customs officer should not be relying only on experience and paperwork.
Imagine a system that can compare:
the declared value of a product;
previous imports by the same company;
prices from the exporting country;
the importer’s historical behaviour;
HS-code patterns;
country-of-origin data;
shipping information;
previous Customs offences;
related companies;
and financial information.
A suspicious shipment could then be automatically flagged.
That would not eliminate fraud.
But it could make the system much harder to game.
The cigarette shipment that shows what Customs is up against
In July 2026, Customs seized approximately 3.6 million cigarettes hidden inside imported cold-storage panels.
More than 18,000 cartons were reportedly concealed inside the panels.
Customs estimated the potential tax loss at more than Rs.400 million, while the market value of the cigarettes was estimated at more than Rs.450 million.
Think about what that means.
The importer did not simply forget to declare a few cartons.
The cigarettes were allegedly concealed inside something else.
This is the other side of the revenue-leakage problem.
Some losses happen because paperwork is wrong.
Others involve deliberate attempts to defeat the inspection system.
And Customs has to distinguish between them.
But there is another problem: what happens after Customs catches you?
This may be one of the most important questions in the entire investigation.
Finding fraud is only half the job.
The State then has to recover the money.
The Auditor General’s 2024 report found that penalties and forfeitures imposed in 184 completed Customs investigations amounted to approximately Rs.4.79 billion.
But that amount had not been recovered as of December 31, 2024.
The 30% share due to the General Treasury from those penalties and forfeitures amounted to approximately Rs.1.437 billion and remained outstanding.
This creates a strange situation.
An investigation can succeed.
An importer can be found liable.
A penalty can be imposed.
And yet the Government can still be waiting for its money.
For the taxpayer, the final question is not:
“Did Customs issue the penalty?”
It is:
“Did the Treasury actually receive the money?”
One jewellery case raises an even bigger question
The Auditor General’s 2024 Annual Report describes a Customs investigation involving jewellery valued at approximately Rs.753.21 million.
The duty payable was approximately Rs.219.82 million.
The audit states that, under the Customs Ordinance, goods could have been confiscated and forfeited at up to three times their value in the circumstances described.
That potential forfeiture was calculated at approximately Rs.2.919 billion.
But at the conclusion of the investigation, the forfeiture was reduced to Rs.190 million.
The importer subsequently released the goods after making a payment that, according to the audit, resulted in an underpayment of approximately Rs.29.82 million compared with the tax amount payable.
Again, this does not establish corruption.
There may be legal reasons for settlements and reductions.
But it raises an important public question:
Who decides when a large Customs penalty should be reduced — and how transparent is that decision?
When billions of rupees are involved, the public deserves to know that such decisions are governed by consistent rules.
The reward system
There is another unusual feature of the Customs system.
Customs operates reward mechanisms for officers and informants connected with seizures and investigations.
The Auditor General found that approximately Rs.25.245 billion had been distributed to Customs officers and Rs.4.554 billion to informants over the 13 years from 2012 to 2024.
The audit questioned the legal and procedural basis for aspects of the reward distribution and recommended an appropriate legal mechanism.
This creates a difficult policy question.
Rewards can encourage officers and informants to detect fraud.
But incentives must also be carefully designed.
If an officer benefits financially from a seizure, the system must ensure that the incentive does not influence how cases are selected, assessed or settled.
The objective should always be:
protecting the revenue.
Not generating rewards.
The old warnings did not begin in 2024
Sri Lanka’s Customs weaknesses are not a new discovery.
Older Auditor General reports have identified problems involving:
container examination;
classification;
valuation;
monitoring;
security cameras;
clearance procedures;
penalties;
bank guarantees;
coordination between government agencies.
A 2016 audit, for example, found deficiencies at a container clearance centre, including concerns about examination procedures, unusually rapid clearance of containers, delays, inadequate monitoring and weaknesses in the use of technology. The audit warned that large-scale revenue losses could not be ruled out.
An earlier audit also documented cases involving erroneous classification and Customs investigations that resulted in significant differences between the revenue originally payable and amounts ultimately recovered.
The important point is not that every old problem remains today.
Some may have been fixed.
Some systems have been modernised.
The important question is:
Which recommendations were actually implemented, and which weaknesses survived?
The problem with celebrating only record revenue
There is a temptation to look at the Rs.2.56 trillion collected in 2025 and the Rs.1.379 trillion collected in the first six months of 2026 and conclude that Customs is working better than ever.
That may be partly true.
The figures are undeniably strong.
But a revenue target is not the same thing as the maximum amount that could legally be collected.
Suppose Customs is supposed to collect Rs.100.
If it collects Rs.110, it has exceeded its target.
But if Rs.120 was actually legally payable and Rs.10 was lost because of undervaluation or enforcement failures, Customs has still achieved its target while the country has lost revenue.
That is why collection efficiency and leakage are two different measurements.
Sri Lanka needs both.
Vehicles show how dependent Customs revenue has become
The current numbers make this particularly obvious.
By June 2026, vehicle imports had generated more than Rs.512 billion in taxes, with motor vehicles accounting for approximately Rs.386.7 billion.
That is a huge contribution.
But it also creates a structural question.
What happens to Customs revenue if vehicle imports fall?
What happens if the Government changes vehicle taxes?
What happens if consumers stop importing because of exchange-rate pressures?
What happens if the country moves toward electric vehicles and the tax structure changes?
A sustainable revenue system should not depend excessively on a narrow group of highly taxed imports.
The ordinary person eventually pays for leakage
Customs revenue sounds like something that happens far away from ordinary life.
It doesn’t.
Government revenue pays for public services.
If Rs.1 billion is legally owed but never collected, that money has to be replaced somehow.
Through another tax.
Through borrowing.
Through spending cuts.
Or through accepting a larger deficit.
That does not mean every rupee of Customs arrears becomes a tax increase.
Government finances are much more complicated.
But the principle is simple:
Revenue that the State should have collected but did not is money that cannot be used for something else.
For a country that went through one of the worst economic crises in its modern history, that matters.
The small importer versus the large importer
There is another human issue that often gets lost.
Sri Lanka’s Customs system does not affect only large corporations.
It affects the small businessman importing spare parts.
The woman bringing fabric for her tailoring business.
The entrepreneur importing electronics.
The restaurant owner importing equipment.
The farmer importing machinery.
The family bringing goods from overseas.
If Customs procedures are slow or unpredictable, these people pay.
Every additional day a container sits at a port can mean additional costs.
Every valuation dispute can delay a business.
Every classification dispute can freeze working capital.
So improving Customs is not simply about collecting more taxes.
It is also about making legitimate trade faster and more predictable.
The danger of over-correction
There is another side to this story.
If Customs responds to leakage by treating every importer as a potential criminal, legitimate businesses suffer.
More inspections mean delays.
More documentation means costs.
More delays mean higher prices.
And higher prices ultimately reach consumers.
That is why modern Customs systems rely on risk-based controls.
The objective should be to identify the suspicious minority without slowing down everyone else.
The Parliamentary Ways and Means Committee’s current attention to technology and inspection procedures is therefore important.
The future of Customs cannot simply be:
“Open more containers.”
It has to be:
“Know which containers to open.”
Where the biggest vulnerabilities appear to be
Based on the documented evidence, several areas deserve particular scrutiny.
1. Import valuation
Undervaluation can directly reduce tax.
Customs itself has created the NIVDU to combat this.
2. Classification
A product placed under the wrong tariff category can attract a different tax.
Older audit reports have repeatedly identified classification-related problems.
3. Bonded warehouses
The 2024 audit’s Rs.7.388 billion vehicle-related revenue gap demonstrates the importance of monitoring these facilities.
4. Penalties and settlements
The jewellery case demonstrates why decisions reducing penalties need strong controls and transparency.
5. Revenue arrears
Billions remain outstanding, including amounts dating back years.
6. Container inspection
Customs cannot inspect everything, making risk assessment and intelligence essential.
7. Inter-agency data
Vehicle registrations, Customs declarations, banking data, company information and tax records can potentially reveal inconsistencies when linked properly.
The technology Customs should be moving toward
Imagine a Customs system where a shipment automatically generates a risk score.
The system sees that:
A company that normally imports Rs.10 million worth of electronics suddenly declares Rs.2 million.
The declared price is 40% below comparable imports.
The exporter is connected to the same beneficial owner.
The importer has previously faced Customs investigations.
The product’s HS classification has changed repeatedly.
The shipment is routed through an unusual intermediary.
The system flags it.
A Customs officer investigates.
That is very different from opening containers randomly.
It is also more difficult to manipulate.
The technology already exists globally.
The challenge is building the institutional infrastructure and connecting Sri Lanka’s databases.
What should Parliament demand?
The Parliamentary Committee on Ways and Means is already examining Customs operations.
But if Parliament genuinely wants to understand revenue leakage, it needs to go beyond asking:
“How much did Customs collect?”
It should ask:
How much was legally payable?
How much was assessed?
How much was actually collected?
How much remains outstanding?
How much was written off?
How much was reduced through settlements?
How many cases involved undervaluation?
How many involved misclassification?
How many containers were physically inspected?
How many were flagged by risk systems?
How many investigations resulted in successful recovery?
And perhaps the most important:
How many Auditor General recommendations have actually been implemented?
The number Sri Lanka does not publish prominently
There is a number missing from the public conversation.
Not Customs revenue.
Not Customs targets.
Not collections.
But:
The estimated value of revenue that Sri Lanka should have collected but did not.
Without that figure, the public can know how much Customs collected.
It cannot easily know how much Customs failed to collect.
That is the difference between measuring performance and measuring leakage.
A serious fiscal-reform programme should eventually publish both.
This is not a story about “bad Customs officers”
That would be too easy.
And it would be unfair.
Thousands of Customs officers perform a difficult job every day.
They inspect cargo.
Detect drugs.
Seize cigarettes.
Stop counterfeit goods.
Collect taxes.
Investigate fraud.
And work in an environment where smugglers have strong financial incentives to defeat them.
The July cigarette seizure is one recent example of that work.
The problem this investigation identifies is broader.
It is about systems.
If a good officer has poor information, the system is weak.
If a good officer cannot access another agency’s data, the system is weak.
If a penalty is imposed but cannot be collected, the system is weak.
If valuation information is incomplete, the system is weak.
If an importer can exploit a classification loophole, the system is weak.
And if the same weaknesses appear year after year in Auditor General reports, the problem becomes institutional.
The uncomfortable conclusion
Sri Lanka Customs is not failing.
The numbers prove that.
It collected a record amount in 2025.
It exceeded its 2026 revenue target by a substantial margin during the first half of the year.
But success at collecting revenue does not prove that leakage has been eliminated.
The Auditor General’s findings show that significant weaknesses remain.
There are billions in arrears.
There are delayed recoveries.
There are cases where the audit calculated substantial differences between taxes that should have been collected and taxes actually collected.
There are questions around penalties.
There are continuing challenges around valuation and inspection.
And there are old recommendations whose implementation needs to be demonstrated.
So perhaps the right question for Sri Lanka is not:
“How much money did Customs collect?”
It is:
“How much money should Customs have collected?”
That is a much harder number.
And perhaps a much more important one.
Because eventually, someone pays
A container leaving Colombo Port does not end the financial story.
The tax collected becomes Government revenue.
The tax that is not collected does not simply disappear.
Someone eventually has to make up the difference.
Sometimes it is another taxpayer.
Sometimes it is a business passing the cost to consumers.
Sometimes it is Government borrowing.
Sometimes it is reduced public spending.
And sometimes the loss is never recovered at all.
For a country that has spent years asking ordinary citizens to accept higher taxes, higher prices and economic sacrifices, there is a powerful principle at stake.
Before asking the public for another rupee, the State should make sure it has collected every rupee that is already legally due.
That does not mean squeezing legitimate businesses.
It does not mean opening every container.
It does not mean treating every importer as a suspect.
It means building a Customs system where honest trade is fast, dishonest trade is difficult, and public revenue is protected.
Sri Lanka has already shown that Customs can collect trillions.
Now the country needs to know something else.
How many billions are still slipping through the gaps?
Lakbima News investigative note
This investigation does not allege that Sri Lanka Customs as an institution is corrupt, nor does it allege that individual Customs officers involved in the cases discussed acted unlawfully. The report distinguishes between revenue actually collected, revenue assessed as payable, outstanding arrears, audit-calculated revenue differences and allegations of deliberate fraud. Auditor General observations are presented as audit findings; where Customs or other authorities have provided explanations or responses, those are identified accordingly.
The Rs.7.388 billion vehicle figure, Rs.29.82 million jewellery underpayment and other figures cited from the Auditor General’s reports are audit findings/calculations and should not be interpreted as proven criminal losses unless a competent authority establishes criminal conduct.
