The Fiscal Endurance of Indonesia’s State Budget in 2026

On 6 April 2026, Finance Minister Purbaya Yudhi Sadewa made a declaration that drew immediate and widespread attention: the prices of subsidised fuels—namely Pertalite petrol and subsidised diesel—would not be raised for the remainder of the 2026 fiscal year. The commitment was delivered before Commission XI of the House of Representatives (Dewan Perwakilan Rakyat/DPR) at a time when global crude oil prices were climbing sharply, driven by an escalation of conflict in the Middle East. The announcement at once raised a fundamental fiscal question: how robust is Indonesia’s State Budget (Anggaran Pendapatan dan Belanja Negara/APBN) in sustaining such a pledge?

This essay presents a comprehensive analysis of the APBN 2026’s fiscal endurance under the strain of swelling energy subsidies. The analysis is structured around three principal dimensions: the initial fiscal position that formed the policy’s point of departure; stress-test scenarios constructed on differing oil price assumptions; and the layered budgetary defences that the government has assembled as shock absorbers.

An Analysis of Budgetary Resilience Amid Mounting Fuel Subsidy Pressures and Global Oil Price Volatility

  

The Initial Fiscal Position and Structural Pressures

The 2026 APBN was designed with total state expenditure of Rp3,786.5 trillion and a revenue target of Rp3,147.7 trillion, yielding a planned deficit of Rp638.8 trillion, equivalent to 2.48 per cent of Gross Domestic Product (GDP). The macro assumption underpinning the entire framework set the Indonesian Crude Price (ICP) at USD 70 per barrel—a figure that proved to be a significant underestimate of actual market conditions.

In the original design, the government allocated Rp210.1 trillion for energy subsidies, comprising Rp25.14 trillion for specific-grade fuel subsidies, Rp80.26 trillion for 3-kilogram LPG cylinders, and Rp104.64 trillion for electricity subsidies. When combined with energy compensation payments, the total subsidy and compensation envelope stood at Rp381.3 trillion. The surge in global oil prices to approximately USD 103 per barrel by early April 2026, however, rendered these figures materially inadequate.

The most telling indicator of this pressure is the budget’s sensitivity to oil price movements: every USD 1 rise in the crude price per barrel adds up to Rp6.8 trillion to the subsidy burden. With the gap between the APBN assumption (USD 70/bbl) and the actual price (approximately USD 103/bbl) standing at USD 33 per barrel, the potential additional exposure across a full fiscal year exceeds Rp200 trillion—a pressure that is no longer marginal in character, but structural.

 

Scenario Analysis: Stress-Testing the APBN

The Ministry of Finance has prepared a series of stress-test scenarios reflecting differing oil price trajectories throughout the year. These scenarios provide a clear picture of the critical thresholds at which the APBN’s resilience begins to fracture.

Optimistic scenario: USD 86 per barrel

At an average oil price of USD 86 per barrel, with the rupiah exchange rate at Rp17,000 per US dollar, projections indicate that the APBN deficit would reach 3.18 per cent of GDP. This already breaches the legal ceiling stipulated in State Finance Law No. 17 of 2003, which caps the deficit at 3 per cent of GDP. Paradoxically, even the scenario labelled ‘optimistic’ would place the government outside its own statutory fiscal corridor. Exceeding this threshold requires parliamentary approval, carrying significant implications for Indonesia’s fiscal reputation in the eyes of international investors and rating agencies.

Moderate scenario: USD 97 per barrel

At USD 97 per barrel with the rupiah at Rp17,300 per US dollar, the deficit is projected to widen to 3.53 per cent of GDP. Significantly, Coordinating Minister for Economic Affairs Airlangga Hartarto stated explicitly that subsidised fuel prices could only be maintained throughout 2026 as long as the average oil price did not exceed USD 97 per barrel. This means the moderate scenario itself effectively represents the upper boundary of the government’s commitment without additional fiscal intervention.

Purbaya’s commitment ceiling: USD 100 per barrel annual average

This is the scenario in which the government has formally pledged to hold subsidised fuel prices. Assuming an annual average of USD 100 per barrel — which implies some months above and some below that figure — the government is targeting a deficit of approximately 2.92 per cent of GDP. This outcome is achievable only if every layer of the budgetary defence mechanism (discussed in Section IV) operates fully and simultaneously. In other words, there is no margin for any single mitigating mechanism to fall short of its target.

Pessimistic and tail-risk scenarios: USD 115–150 per barrel

At USD 115 per barrel with the rupiah at Rp17,500 per US dollar, the deficit is projected to reach 4.06 per cent of GDP, equivalent to approximately Rp1,004 trillion. At this level, the government’s debt ratio could approach or exceed 45 per cent of GDP, narrowing monetary policy manoeuvre space and substantially increasing debt servicing costs. At the extreme tail of USD 150 per barrel — a scenario Purbaya himself has acknowledged—the Saldo Anggaran Lebih (SAL) fiscal reserve becomes the only remaining instrument of defence. Under such conditions, the APBN would no longer be able to stand on its own structural footing.

 

Six Layers of Fiscal Defence

The government is not relying solely on the existing budget posture. Its fiscal defence strategy is deployed in layers, with each serving as a buffer before the next is activated.

The first layer is the energy subsidy allocation within the APBN itself (Rp210.1 trillion), constituting the primary line of defence. The second is expenditure efficiency across ministries and government agencies, implemented in stages by trimming budget lines deemed less of a priority. The third layer draws on windfall gains from Non-Tax State Revenue (PNBP) in the energy and mineral resources sector: paradoxically, elevated oil and coal prices simultaneously boost state revenues, a portion of which can be redirected to offset the very subsidy burden they create. The fourth layer involves PT Pertamina acting as a temporary absorber, whereby the state-owned enterprise initially bears the price differential before receiving government compensation of approximately 70 per cent per month on a rolling basis.

The fifth layer is the B50 biodiesel programme, scheduled to take effect on 1 July 2026. By reducing dependence on fossil-based fuel by approximately 4 million kilolitres, the programme is projected to cut the energy subsidy burden by up to Rp48 trillion — a saving of considerable significance. The sixth and final layer is the SAL reserve of Rp420 trillion, held at Bank Indonesia (Rp120 trillion) and a consortium of state-owned banks or Himbara (Rp300 trillion). The SAL represents an accumulation of unspent budget surpluses from previous fiscal years, and Purbaya has stated unequivocally that this instrument will only be activated under genuinely pressing circumstances, and only with the consent of Commission XI of the DPR.

 

Critical Pressure Points and Medium-Term Risks

Whilst the APBN can technically sustain the fuel price freeze through end-2026 under the government’s central commitment scenario (USD 100/bbl annual average), several risks demand careful scrutiny.

First, exchange rate risk functions as an amplifier. Every Rp100 depreciation in the rupiah per US dollar does increase state revenues by approximately Rp5.3 trillion, but it simultaneously raises state expenditure by Rp6.1 trillion — yielding a net deficit-widening effect. Should rupiah depreciation coincide with a surge in oil prices, the cumulative impact would be considerably more severe than the arithmetical sum of either factor in isolation.

Secondly, there is a high degree of dependence on the simultaneous success of all defensive layers. The government’s target of a 2.92 per cent deficit at USD 100/bbl is only realisable if ministry expenditure efficiency, PNBP windfall revenues, the B50 programme, and Pertamina’s absorptive capacity all perform in accordance with their respective targets. Failure in any single one of these mechanisms would push the deficit dangerously close to the legal 3 per cent ceiling.

Thirdly, the fiscal position as at end-March 2026 already shows a deficit of Rp240.1 trillion, or 0.93 per cent of GDP, within the first three months of the fiscal year alone. Should this pace of expenditure continue — and subsidy pressures tend to intensify rather than ease as oil prices rise — the annual deficit target will come under increasing strain. Tax revenue growth of 20.7 per cent year-on-year provides a degree of relief, but the 3 per cent decline in PNBP, attributable to the rerouting of state-owned enterprise dividends to the Danantara investment vehicle, is a signal that warrants vigilance.

Fourthly, under the pessimistic scenario (USD 115/bbl), even the Rp420 trillion SAL would prove insufficient to bridge the deficit gap on its own. At a deficit level of 4.06 per cent of GDP (Rp1,004 trillion), the gap between the actual deficit and a fiscally sustainable position exceeds what the SAL can patch without fundamentally compromising its function as a long-term national reserve.

 

Conclusion

Indonesia’s 2026 APBN finds itself in a position that is technically manageable, yet genuinely precarious. The government’s commitment not to raise subsidised fuel prices is a pledge that can be honoured through end-2026 — but only under one very specific and demanding set of conditions: that the average annual crude oil price does not exceed USD 100 per barrel, that the rupiah exchange rate remains relatively stable, that every mitigation programme performs as designed, and that no additional external shocks arrive to further erode the fiscal position.

Put differently, the APBN 2026 possesses sufficient resilience to complete this fiscal year without reneging on its subsidy commitment, but with virtually no margin for error. The Rp420 trillion SAL is a formidable last line of defence — but it is not an inexhaustible one. In the medium term, absent a more targeted subsidy reform and a broadened national revenue base, the policy of suppressing fuel prices in an era of geopolitical instability will continue to place Indonesia’s public finances on an ever-narrowing ledge each time global oil markets are thrown into turmoil.

The most important lesson of this analysis is not whether the APBN will hold—but rather the price that must be paid for that resilience, and whether that price is commensurate with the social protection it affords to the population it is designed to serve.

Key fiscal parameters (APBN 2026)

Original subsidy & compensation budget

Rp381T

Baseline allocation

Additional subsidy burden

+Rp90–100T

Due to oil price shock

Total projected subsidy outlay

~Rp481T

Revised upward

SAL fiscal buffer

Rp420T

Emergency reserve

Sensitivity per USD $1/bbl rise

Rp6.8T

Additional subsidy cost

APBN deficit at end-March 2026

Rp240T

0.93% of GDP

Stress-test scenarios — fiscal endurance by oil price

ScenarioOil price (ICP)Deficit/GDPVerdict

APBN baselineUSD 70/bbl2.48–2.68%Safe

OptimisticUSD 86/bbl~3.18%Watch

Moderate (Gov't ceiling)USD 97/bbl~3.53%Stretched

Purbaya's commitment ceilingUSD 100/bbl avg~2.92% (targeted)Manageable

PessimisticUSD 115/bbl~4.06%Critical

Extreme tailUSD 150/bbl>4.5% est.SAL required

Legal ceiling under UU Keuangan Negara No. 17/2003 is 3% of GDP. Breaching it requires parliamentary consent. Under the optimistic scenario, this ceiling is already breached.

Layers of fiscal defence — stacked buffers

Layer 1 — existing APBN subsidy allocation (Rp210T energy subsidi)

Layer 2 — ministry & agency (K/L) expenditure efficiency savings

Layer 3 — PNBP windfall from rising commodity prices (oil, coal)

Layer 4 — Pertamina absorbs short-term via monthly compensation (70%/mo)

Layer 5 — biodiesel B50 rollout (July 2026, saves ~Rp48T)

Layer 6 — SAL emergency reserve (Rp420T — last resort)

Layers are deployed sequentially. The government has stated SAL will only be activated if Layers 1–5 prove insufficient. Each layer narrows the fiscal gap, reducing dependence on the reserve.

Endurance timeline — how long can the APBN hold?

Sources: Ministry of Finance of the Republic of Indonesia; Working Session of Commission XI, DPR RI (6 April 2026); ANTARA; Kompas.id; CNBC Indonesia; VIVA; Tempo.

Bahasa

War : Survivors, Memory, and Moral Responsibility (30)

In the theatre of international relations, few figures have disrupted the established script as provocatively as Donald Trump. His "America First" doctrine, often perceived as a paradoxical blend of isolationism and aggressive unilateralism, has forced a global re-evaluation of the American hegemon's intentions. While the drums of conflict often beat to the rhythm of "stability" and "democracy," the underlying cadence frequently suggests a more pragmatic, perhaps even more cynical, choreography of power.

The central enigma of this era lies in the true catalyst behind the escalations and military posturing witnessed during his tenure. One must consider whether these confrontations were primarily driven by a neo-mercantilist desire to secure foreign oil reserves, thereby continuing a long-standing tradition of using energy as a formidable lever of geopolitical influence. Alternatively, were these actions intended to act as a bellicose stimulus for the American economy, seeking to guarantee domestic prosperity through a policy of high-stakes external aggression?

Furthermore, there is the compelling possibility that such conflicts were merely components of a more expansive ambition: a relentless projection of dominance designed to reassert the United States as the world's uncontested superpower. Yet, as with all complex geopolitical manoeuvres, the surface narrative rarely tells the whole story. Could there be other, more clandestine motives—hidden beneath the layers of populist rhetoric and strategic posturing—that provide the definitive explanation for these global confrontations?

Commentators often note that military interventions or confrontational policies under Trump have been framed in several ways. Some argue they were linked to securing energy resources, particularly oil, given the long-standing role of energy in U.S. foreign policy. Others suggest they were primarily aimed at bolstering American economic growth by ensuring favourable trade conditions and protecting strategic industries. A further perspective is that such actions were part of a broader attempt to project U.S. power globally, reinforcing dominance in international affairs.

In short, interpretations vary: some see oil as the driver, others see economic protectionism, and still others see a bid for global influence. What unites these views is the recognition that energy, economics, and geopolitical ambition are deeply intertwined, and that no single explanation fully captures the complexity of U.S. actions.

However, speculation regarding conventional motives such as oil and military prowess in the Trump era may merely represent the final chapter of an old manual on power. Today, the global paradigm has shifted radically, where influence is no longer measured solely by the count of warheads or the vastness of seized oil fields. We now stand at the threshold of a new epoch that redefines sovereignty, compelling us to pose a question far more critical to the future of civilisation.

In this increasingly complex and digitised global landscape, who truly holds the key to world dominion? Will the global hierarchy continue to be dictated by the oil magnates who control physical energy, or has that control transitioned into the hands of the economic titans who orchestrate the flow of global capital? Alternatively, could the throne of future power be occupied by the masters of AI who govern artificial intelligence and algorithms, or perhaps by those most adept at controlling the narrative and shaping public perception within the digital realm?

In our present age, the notion of who might truly “rule the world” is less about crowns or armies and more about the forces that shape everyday life. Artificial intelligence, for instance, has the capacity to infiltrate nearly every sphere—commerce, defence, culture, even governance—yet its power remains tethered to regulation, ethics, and public trust. Economic giants, meanwhile, continue to wield immense influence, able to sway governments and dictate the rhythms of global trade, though their dominance is often fragile in the face of crises. Oil, once the undisputed lever of geopolitics, is now challenged by the urgency of climate change and the rise of renewable energy, though control of energy resources still confers formidable leverage.

And then there is narrative. In an era defined by social media and instantaneous communication, those who can craft and disseminate compelling stories hold sway over hearts and minds. Narratives legitimise power, galvanise movements, and destabilise regimes. They are intangible yet pervasive, capable of reshaping reality itself.

If one were to distil it, artificial intelligence provides the infrastructure and capability, while narrative supplies the legitimacy and direction. Economics and energy remain vital, but without a persuasive story or technological scaffolding, their grip weakens. In truth, the contest is not between these forces in isolation, but in how they intertwine: algorithms that amplify narratives, economies that depend on technological systems, and energy that fuels both.

So the question becomes less “who rules” and more “which combination of forces will define the future.” Would you find it more unsettling to live under the cold precision of algorithms, or under the emotional sway of narratives that bend perception itself?

A world steered by narrative is indeed frightening, for it can infiltrate collective consciousness and bend emotions with subtlety. Yet the dominion of cold algorithms presents a different, perhaps deeper, unease. These systems no longer merely follow explicit instructions; they learn, adapt, and generate outcomes that even their creators cannot fully anticipate. This opacity, often described as the “black box” problem, means that while we can witness the results, the reasoning behind them remains shrouded.

The true peril lies not simply in our inability to predict, but in our growing dependence upon such systems. As more aspects of life—finance, security, health, even social interaction—are mediated by algorithms, we risk surrendering control to mechanisms we do not comprehend. Narratives may deceive the heart, but algorithms can quietly restructure reality itself. When the two converge—algorithms amplifying narratives—the potential for unforeseen transformation becomes profound.

Thus, the tension is stark: the unpredictability of algorithms versus the deliberate certainty of manipulative narratives. One unsettles because it cannot be fathomed, the other. After all, it is designed to mislead. Which of these, do you think, would ultimately prove the more corrosive to human freedom?

Narratives, however manipulative, still leave traces of their origin. They can be challenged, countered, and dismantled by other narratives. Algorithms, on the other hand, are far more elusive. They operate beneath the surface, hidden in layers of code and data, producing outcomes that often appear neutral or inevitable. Yet the logic that drives them is rarely transparent, and even those who design them may struggle to explain their decisions once the systems begin to learn and evolve.

This opacity is precisely what makes algorithmic power so unsettling. Unlike a narrative, which can be debated in the public square, an algorithm can quietly shape reality without revealing its hand. It decides what information you see, which opportunities you are offered, and even how institutions treat you—all while cloaked in the aura of objectivity. The danger lies in the fact that we cannot easily contest what we cannot see or understand.

In that sense, the fear of algorithms is not only about unpredictability, but about invisibility. They do not merely manipulate emotions; they restructure the very conditions under which emotions, choices, and narratives emerge. And when such systems are entrusted with decisions that affect millions, the question of who truly holds power becomes far more difficult to answer.

The question of whether human beings can truly control algorithms is a vexed one. In principle, algorithms are human creations, designed, coded, and deployed by us. Yet once they are set loose in the world, particularly those built upon machine learning, they begin to evolve in ways that exceed our immediate grasp. They absorb vast amounts of data, identify patterns invisible to human perception, and generate outcomes that may surprise even their architects. This does not mean they are entirely beyond control, but rather that the nature of control shifts: it becomes less about direct command and more about governance, oversight, and the imposition of ethical boundaries.

Humans can regulate the conditions under which algorithms operate, limit their scope, and demand transparency. We can insist upon accountability, requiring that those who deploy such systems explain and justify their use. Yet the deeper challenge lies in the opacity of the systems themselves. To “control” an algorithm in the fullest sense would mean understanding and anticipating every decision it makes, and that is increasingly impossible. What remains within our reach is the ability to decide where algorithms may be applied, how their outputs are interpreted, and whether their authority is accepted.

Thus, the question is not whether humans can control algorithms in the mechanical sense, but whether we can control the social and institutional frameworks that give them power. The danger is not that algorithms act alone, but that human institutions abdicate responsibility, hiding behind the supposed neutrality of machine logic. In that abdication lies the greatest risk: not the algorithm itself, but our willingness to let it govern without challenge.

The gravest danger does not reside in the algorithms themselves, but in the human institutions that choose to cloak their authority behind them. Algorithms, after all, are tools—complex, opaque, and often unpredictable, but tools nonetheless. It is the decision to elevate them to the status of unquestionable arbiters that transforms them into instruments of domination. When governments, corporations, or other centres of power invoke the neutrality of machine logic, they absolve themselves of responsibility, presenting outcomes as inevitable rather than chosen.

This abdication of accountability is what renders the situation so perilous. The algorithm becomes a convenient mask, a way of enforcing decisions without debate, of silencing dissent by appealing to the supposed objectivity of code. In truth, every algorithm reflects human choices: the data selected, the objectives defined, the contexts in which it is deployed. To pretend otherwise is to surrender agency and to allow power to operate without scrutiny.

Thus, the true threat lies not in the machinery itself, but in the willingness of institutions to hide behind it, to wield its authority while disavowing their own. In that concealment, the line between governance and manipulation blurs, and the possibility of resisting or contesting decisions diminishes. It is not the coldness of the algorithm that should frighten us most, but the warmth of human hands that place it upon the throne and then retreat into the shadows.

The answer is not that a single force—be it artificial intelligence, economic might, oil, or narrative—will reign supreme, but rather that power in our age is defined by the interplay between them. Artificial intelligence provides the infrastructure, the machinery through which decisions are made and realities are shaped. Economic power continues to dictate the distribution of resources and the stability of nations. Oil and energy, though challenged by climate imperatives, remain critical levers of influence. Yet it is narrative that confers legitimacy, that persuades populations to accept or resist, that transforms raw power into authority.

The greatest danger lies not in any one of these forces in isolation, but in the way human institutions wield them together. An algorithm may be cold, but it is human hands that place it upon the throne. A narrative may be manipulative, but it is human voices that choose to spread it. Oil and money may command, but it is human systems that decide how they are deployed. Thus, the true rulers of the world are not the tools themselves, but those who hide behind them, cloaking their authority in the neutrality of machines, the inevitability of markets, or the seduction of stories.

In this sense, the world is most likely to be ruled by those who master the fusion of technology and narrative, institutions that can harness algorithms to amplify stories and use those stories to justify economic and political power. It is not AI alone, nor narrative alone, but the convergence of both, wielded by human institutions that refuse accountability, that poses the most formidable claim to global dominion.

Key References

Abhijeet Sarkar, AI and Global Power Shifts: The New Geopolitical Battleground, 2025, Synaptic AI Lab Press 

Sarkar explores how artificial intelligence is redefining geopolitics, arguing that AI is becoming a decisive factor in global power rivalries, surpassing traditional military and economic dominance.

Nick Srnicek, Silicon Empires: The Fight for the Future of AI, 2026, Polity Press 

Srnicek examines the struggle among corporations and states to control AI, emphasising how narratives around technology are deployed to legitimise authority and reshape societies.

IEEE Collective, The Power of Artificial Intelligence for the Next-Generation Oil and Gas Industry, 2024, IEEE eBooks 

This work demonstrates how AI is transforming the energy sector, showing the convergence of digital technologies with the oil and gas sector, and how control of energy resources remains central to geopolitical leverage.

[Part 31]

[Part 29]

Bahasa