Research Brief: BI Working Paper — At B50–B70, Lost CPO Exports Outweigh Diesel Import Savings

Eleven Bank Indonesia researchers conclude that at blends from B50 to B70, the crude palm oil (CPO) export earnings Indonesia forgoes systematically exceed the savings on diesel imports. As a result, the current account deficit widens and real GDP and household consumption fall. The findings appear in the working paper "Pushing Biodiesel Boundaries: B50 & Beyond – Promise or Premature?" (WP/05/2025) by MHA Ridhwan and colleagues, posted on Bank Indonesia's website in March 2026. The authors stress that the paper reflects their own views, not Bank Indonesia's official position.

Key findings

• External balance. In the simulation, the additional current account deficit caused by the biodiesel programme rises from 0.18% of GDP in 2025 to 0.32% of GDP in 2030 if the blend is held at B50. It reaches 0.38% of GDP under a gradual increase and 0.45% of GDP on an ambitious path towards B70.
• Growth and welfare. The computable general equilibrium (CGE) model shows real GDP on the expenditure side falling 0.54% (B50), 0.71% (B60) and 0.78% (B70) against the baseline. Real household consumption falls 0.19%, 0.45% and 0.47%, while import volumes rise 1.99%, 2.2% and 2.71%.
• Palm-producing regions. Under B50, regional GDP falls below the baseline in all 10 leading palm oil provinces. The steepest declines are in East Kalimantan (−0.62%), Central Kalimantan (−0.53%) and South Sumatra (−0.44%).
• Incentive burden. In several periods the biodiesel incentive exceeded IDR 6,000 per litre, while the diesel subsidy was only about IDR 500–1,500 per litre. As long as the per-litre incentive is larger than the diesel subsidy, every increase in the blend directly raises funding needs.
• Inflation. The effect of higher CPO prices on inflation under B50 and B60 is estimated at only around 0.01–0.02 percentage points.
• Emissions versus land. The simulation finds that B50 cuts fuel emission intensity by 32.20% (gCO₂e/MJ) compared with pure fossil diesel. However, B50 requires 17.85 million kL of biodiesel, 21 million tonnes of CPO and 5,250,000 hectares of land. The authors warn that if the extra feedstock comes from converting forest and peatland, the emission savings could be eroded or even reversed.

Methodology

The authors build material balance sheets for CPO and biodiesel. From these they calculate the difference between diesel import savings and forgone CPO export earnings, and estimate the subsidy needed under each scenario. The results feed into a 207-sector CGE model covering 34 provinces, which combines the ORANI and WAYANG models and is calibrated on Statistics Indonesia's national and interregional input-output tables. The size of each shock comes from SARIMA projections of monthly data for January 2021 to August 2025, with B30 as the baseline. For B50, diesel output falls 38% and biodiesel output rises 27%. The figures are −46% and +33% for B60, and −55% and +39% for B70. The study also draws on focus group discussions with Pertamina Patra Niaga, the Ministry of Energy and Mineral Resources (ESDM), LPEM FEB UI, BRIN, the palm fund agency (then BPDPKS), the Indonesian Biofuel Producers Association (APROBI) and INDEF, as well as a field survey at the Balongan fuel terminal.

Limitations

• Stated by the authors: the paper is research in progress and does not represent Bank Indonesia's position.
• Stated by the authors: Armington, labour and export-demand elasticities are set uniformly across sectors, so the model cannot capture sector-specific responses.
• Stated by the authors: the assumption of constant palm productivity is a "strong" one. If yields fall short, the pressure on exports and non-energy uses would be greater. The authors also call for further work linking energy modelling with land-use change and peat emission dynamics.
• Editor's note: the model data end in August 2025 and use a B30 baseline, before B50 was actually introduced in July 2026. The current account table gives no nominal unit, so we cite only the ratios to GDP. In addition, the text puts the inflation effect at 0.01–0.02 percentage points, while the appendix table totals 0.0402 (B50) and 0.0805 (B60) percentage points.

Relevance for B50/B60 policy

The authors recommend treating any move beyond B50 as a conditional target. The conditions are demonstrable gains in CPO productivity, feedstock diversification, a sound financing architecture and land-use governance. In the short term, they propose a comprehensive impact assessment before any blend increase, plus a flexible blending mechanism that adjusts to movements in diesel, biodiesel and CPO prices. The paper's B60 scenario is relevant because the government is now studying B60 for 2027.

The government's emphasis is different. Eniya Listiani Dewi, Director General of New, Renewable Energy and Energy Conservation (EBTKE), was quoted by ANTARA as saying B50 is projected to save up to IDR 157.28 trillion in foreign exchange in 2026. That figure covers import savings, while the BI paper argues such savings should be weighed against forgone CPO exports. A study by Pranata UI, reported by InfoSAWIT, points in a similar direction: it said B50 could save up to IDR 172.35 trillion in diesel import spending but risks cutting CPO exports by IDR 190.5 trillion.