Yesterday, thousands of Sri Lankan protesters entered the official residence of President Gotabaya Rajapaksa.
Videos showed people walking through the presidential complex, sitting inside rooms, gathering around the grounds and swimming in the president’s pool. Elsewhere in Colombo, protesters entered other government buildings. Prime Minister Ranil Wickremesinghe announced that he was prepared to resign, and parliamentary leaders subsequently said Rajapaksa had agreed to step down on Wednesday.
The images are extraordinary, but they can make the crisis appear more sudden than it actually is.
Sri Lanka has been moving towards this moment for months. The country is experiencing its worst economic crisis since independence. Foreign-exchange reserves have collapsed. Fuel has become extraordinarily difficult to obtain. Medicines and other imported goods are scarce. Inflation has risen sharply. Electricity shortages have disrupted everyday life. The government suspended payments on much of its foreign debt earlier this year and is negotiating with the International Monetary Fund.
The immediate political conclusion seems straightforward: the economy failed, so people turned against the government.
That is true.
It is not enough.
Countries experience recessions, inflation and debt crises without protesters occupying presidential residences. Economic hardship creates dissatisfaction, but whether dissatisfaction becomes a crisis of the political system depends on how citizens understand responsibility.
Sri Lanka’s crisis became particularly politically explosive because many citizens do not view the collapse as an unavoidable external shock. They associate it with decisions made by the government and, increasingly, with one political family.
The Rajapaksas have dominated Sri Lankan politics for much of the last two decades.
Mahinda Rajapaksa served as president from 2005 until 2015 and returned as prime minister in 2019. His younger brother Gotabaya became president that year. Other members of the family occupied senior government positions, including Basil Rajapaksa as finance minister and Namal Rajapaksa in the cabinet.
This concentration of power matters because it concentrates responsibility too.
When political authority is distributed across several parties and institutions, governments can blame coalition partners, predecessors or external constraints.
When one family occupies many of the state’s most important offices, the public has a much clearer target.
The Rajapaksas built enormous political legitimacy partly around the end of Sri Lanka’s civil war. In 2009, government forces defeated the Liberation Tigers of Tamil Eelam, ending nearly three decades of conflict.
For many members of Sri Lanka’s Sinhalese majority, Mahinda and Gotabaya Rajapaksa became associated with national security, victory and strong leadership.
That legitimacy was powerful.
It was not permanent.
This is a pattern I keep returning to: historical achievement depreciates politically.
A government can derive legitimacy from winning a war, achieving independence, defeating terrorism or producing rapid economic growth.
Eventually a new generation asks a different question.
What are you doing now?
Sri Lankans waiting days for petrol are unlikely to be satisfied indefinitely by arguments about what the Rajapaksas accomplished thirteen years ago.
Political legitimacy has to be renewed through performance.
That is particularly true when leaders centralise power around themselves.
Personalised government creates a bargain.
The leader says: give me greater authority because I can deliver stability and results more effectively than fragmented institutions.
If the results arrive, centralisation can become popular.
If they do not, the same concentration of authority makes it difficult to redirect blame.
Strongman politics therefore contains an asymmetry.
It concentrates credit during success.
It concentrates responsibility during failure.
Sri Lanka is experiencing the second half.
Several policy decisions have become central to criticism of the government. Shortly after Gotabaya Rajapaksa took office, his administration introduced substantial tax cuts, reducing government revenue at a time when public finances were already vulnerable.
Then came COVID-19.
Sri Lanka depends heavily on tourism for foreign currency. The pandemic devastated international travel and therefore damaged one of the country’s major sources of dollars.
That external shock was severe and should not be minimised.
No Sri Lankan government created COVID.
But economic resilience is partly about whether governments possess enough fiscal and external buffers to absorb shocks they did not create.
Sri Lanka entered the pandemic with significant debt vulnerabilities.
The collapse in tourism exposed them.
Then policy mistakes compounded the problem.
One of the most controversial was the government’s sudden 2021 ban on chemical fertiliser imports, introduced as part of an attempt to move agriculture rapidly towards organic production while conserving foreign currency.
The objective may have contained legitimate environmental concerns.
The implementation was disastrous.
Farmers had little time to adapt. Agricultural output was disrupted, including production of rice and tea. The policy was later reversed, but the economic damage and loss of confidence remained.
This is a useful example of the difference between policy intention and state capacity.
A government can announce an ambitious objective.
Whether the economy can implement it depends on institutions, timing, technology and incentives.
Moving an agricultural system away from chemical fertilisers may be possible over time.
Doing it almost immediately through prohibition creates a completely different adjustment problem.
Again, transition costs matter.
Policy cannot simply legislate away constraints.
The same mistake appears across governments when political leaders confuse issuing an order with producing an outcome.
Ban something and assume substitutes will appear.
Set a target and assume capacity will follow.
Fix a price and assume supply will remain.
Government possesses legal authority.
It does not possess authority over physical scarcity.
Economics eventually reasserts itself.
Sri Lanka’s foreign-exchange crisis demonstrates this brutally.
The country needs dollars to purchase fuel, medicine and other imports.
If foreign-currency inflows collapse while external debt obligations remain large, the government cannot solve the problem simply by ordering more imports.
Someone has to provide foreign currency.
This is why balance-of-payments crises can become politically destabilising so quickly. They convert abstract macroeconomic imbalances into physical shortages.
A fiscal deficit may initially appear as a number in a government budget.
A foreign-exchange shortage appears as a queue outside a petrol station.
The second is politically much more visible.
Sri Lanka’s inflation compounds the effect. Higher prices reduce real incomes, but shortages create an even deeper sense that ordinary economic systems have stopped functioning.
Money becomes less useful when the product itself is unavailable.
This changes political psychology.
Citizens are not simply poorer.
They experience the state as incapable of providing the basic conditions under which everyday life operates.
That is where an economic crisis becomes a state-capacity crisis.
Can hospitals obtain medicines?
Can buses obtain diesel?
Can power stations obtain fuel?
Can farmers obtain inputs?
Can businesses import materials?
Once the answer repeatedly becomes no, citizens stop judging the government only through ideology.
They begin judging whether government itself works.
This is why yesterday’s protests should not be interpreted simply as left versus right or one party against another.
The protest movement, often referred to through the slogan “Gota Go Home”, has drawn support from Sri Lankans with different political backgrounds united primarily around the demand that Rajapaksa leave.
Negative coalitions can become extraordinarily powerful during crises because they postpone disagreements about what should replace the government.
Everyone can agree the president should go.
The harder question begins the morning after he does.
Who governs?
What happens to the debt?
How quickly should subsidies be reduced?
What taxes should rise?
How should losses be distributed between creditors, taxpayers and households?
What conditions will the IMF require?
How can social protection be maintained while government finances are stabilised?
Those questions will divide the coalition that protest has temporarily united.
This is the transition from opposition to governance.
It is where revolutionary or protest movements frequently struggle.
Removing a leader changes who makes decisions.
It does not change the constraints surrounding those decisions.
Sri Lanka will still need foreign currency after Rajapaksa leaves.
Oil will still need to be imported.
Debt will still need restructuring.
Government revenue will still need to rise.
Inflation will still need to be controlled.
The political system can change overnight.
The balance of payments cannot.
That distinction should temper expectations.
It should not become an argument for keeping an unpopular government simply because alternatives face the same economic constraints.
Accountability still matters.
If leaders made serious policy mistakes, citizens have every right to demand political consequences.
But replacing leaders and solving problems are separate achievements.
Confusing them can produce another cycle of disappointment.
The IMF will probably become increasingly important, which introduces another political difficulty. Sri Lanka needs external financing and a credible path towards debt sustainability. An IMF programme can help restore confidence and unlock additional financing.
But IMF assistance normally involves conditions.
Taxes may need to rise.
Loss-making state enterprises may need reform.
Energy prices may need to reflect costs more accurately.
Government spending may face constraints.
These policies can improve fiscal sustainability while imposing immediate pain on citizens who have already experienced enormous hardship.
This is where economic necessity collides with political legitimacy.
An adjustment programme can be technically correct and politically impossible if the distribution of costs appears unfair.
If ordinary households are asked to accept higher electricity prices while politically connected elites appear protected, reform will lose legitimacy quickly.
The sequencing therefore matters.
Who pays first?
Who is protected?
Which assets are taxed?
How are poorer households compensated?
How transparently are losses distributed?
Economic stabilisation is partly a distributional negotiation.
This connects directly to what I wrote about climate policy and sanctions. Governments can often identify an aggregate policy that improves the economy overall.
Politics determines who bears the transition.
Fairness therefore becomes part of effectiveness.
A reform programme that collapses politically does not produce long-term macroeconomic stability.
Sri Lanka also illustrates the danger of family-based political power. Political dynasties exist in many democracies, including India, Pakistan, the Philippines and the United States.
A famous surname does not automatically make a politician illegitimate.
Voters can freely choose members of the same family.
The institutional problem begins when family relationships become pathways through which political, administrative and economic authority are concentrated beyond ordinary accountability.
Families possess unusually high internal trust.
That can make them attractive to leaders who fear betrayal.
A president may trust a brother with finance more than an independent technocrat or rival politician.
But what increases personal trust can reduce institutional quality.
Appointments become based partly on loyalty rather than comparative competence.
Information becomes less independent.
Officials may become reluctant to challenge decisions made by powerful relatives.
Accountability becomes harder because institutions overlap with family relationships.
This is one reason strong states attempt to make offices impersonal.
The finance ministry should function regardless of who the president’s brother is.
The central bank should produce uncomfortable information even when politicians dislike it.
The civil service should retain expertise across governments.
Institutionalisation means the state becomes less dependent on personal relationships.
Sri Lanka’s crisis demonstrates the risks when citizens begin perceiving the opposite.
There is also a question about corruption, although claims should be handled carefully. Protesters frequently accuse the Rajapaksa family and political elite of corruption and mismanagement.
Some allegations are well documented; others remain political accusations.
We should not use economic collapse itself as proof that every corruption allegation is true.
But perception matters independently.
If citizens believe elites enriched themselves while the country became unable to import fuel, austerity becomes much harder to legitimise.
People accept sacrifice more readily when they believe sacrifice is shared.
This is procedural fairness again.
The president’s residence becoming a symbol yesterday is therefore unsurprising.
A luxurious state residence occupied by protesters during a national shortage communicates inequality more powerfully than a fiscal table ever could.
Political movements rely on symbols because symbols compress complex arguments into images.
A swimming pool inside the presidential residence becomes a representation of elite privilege while ordinary people queue for fuel.
Whether that symbolism captures the full economic explanation is almost irrelevant politically.
It captures the emotion.
The constitutional process now matters enormously.
Rajapaksa has reportedly agreed to resign on Wednesday.
If he does, Sri Lanka needs to transfer power through constitutional procedures rather than allowing the collapse of one administration to become the collapse of the state.
This distinction is essential.
Protest can remove legitimacy from a leader.
It cannot replace the institutions required to govern forty-eight hours later.
Parliament therefore becomes critical.
Sri Lanka needs a government capable of negotiating with creditors, the IMF, India and other partners while retaining enough domestic legitimacy to implement painful reforms.
That may require a broad coalition or temporary national government.
Whatever arrangement emerges, the central objective should be preventing political vacuum.
The comparison with Libya is useful. Removing Gaddafi destroyed a regime faster than institutions could replace it, contributing to prolonged fragmentation.
Sri Lanka possesses much stronger state institutions and a completely different political context, so there is no reason to expect the same trajectory.
But the general principle remains: removing a government and preserving government are different tasks.
Successful protest should aim to change who controls institutions, not destroy the institutions necessary for collective action.
This is where yesterday’s dramatic images need careful interpretation.
People entering the president’s residence can look like the state collapsing.
It may instead be evidence that one government’s authority has collapsed while the broader constitutional state remains capable of transition.
The next several days will tell us much more.
As of tonight, we should not pretend the outcome is settled.
Rajapaksa has said through parliamentary leaders that he will resign.
He has not yet formally left office.
Wickremesinghe has offered to resign.
Parliament still has to determine what follows.
The responsible position is therefore uncertainty.
But the political lesson is already substantial.
Economic crises become regime crises when citizens stop believing hardship is temporary, unavoidable and fairly distributed.
Sri Lanka experienced several genuine external shocks.
COVID destroyed tourism.
Global energy and food prices have risen.
Russia’s invasion of Ukraine worsened commodity pressures.
Those factors matter.
They do not absolve domestic policy.
Good government is partly the capacity to absorb shocks without allowing them to become catastrophes.
That requires reserves, credible fiscal policy, competent institutions and the willingness to reverse mistakes quickly.
Sri Lanka lacked enough of those buffers.
The Rajapaksas built political legitimacy around strong leadership.
The same concentration of power that strengthened their image during successful periods now concentrates public anger around them.
Yesterday, that anger entered the president’s home.
It would be easy to treat the image as the end of the story.
It is probably the beginning of the harder part.
Protest can answer who should leave.
Economics will now demand answers to what should change.
And the second question is considerably more difficult.
