Audit log · research pass 1

Every search, every figure, every derivation

The working file behind the analysis: the originating question, every search run, every figure extracted with its source, and every derived calculation. Published so that any number on this site can be traced back or shown to be wrong.

Question: How will the September 2026 fuel price spike (Tk 145 → 165 octane; diesel Tk 115 → 135) affect Bangladesh? Date of pass: 21 September 2026 (day the hike took effect) Purpose: Complete audit trail — the originating conversation, every search run, every figure extracted, its source, and every derived calculation. This is the working file behind impact_analysis.md.

Verification status legend


Part 1 — The originating conversation

This pass began from a prior conversation (another assistant) that the user pasted in. Recorded verbatim in substance, because Pass 1's first job was to check its claims.

1.1 User's opening prompt

"Do an analysis of fuel cost vs avg income of every country"

1.2 Prior assistant's response — global fuel affordability framework

Defined fuel affordability as the percentage of average monthly income needed for a 50-litre tank, and argued absolute pump price does not determine pain. Four archetypes:

# Archetype Pump price 50L as % of monthly income Examples
1 High income, high fuel cost (European/Nordic) $1.90–2.35/L 2.0–3.5% Norway, Denmark, Germany, Netherlands, UK
2 High income, low/moderate cost (US & Gulf) $0.50–1.15/L 0.8–1.8% USA, Switzerland, Qatar, UAE, Saudi, Australia
3 Low/middle income, subsidised pump (State-Buffer) $0.05–0.75/L 5–15% Egypt, Algeria, Iran, Venezuela, Malaysia, Indonesia
4 Low/middle income, market/taxed pump (High Fuel Stress) $0.95–1.30/L 25–60%+ India, Pakistan, Nigeria, Kenya, Madagascar, Philippines

Affordability matrix it provided (all figures [UNVERIFIED], described by the source itself as estimates):

Country Fuel price ($/L) 50L tank ($) Avg monthly net wage ($) % of income per 50L
United States 0.95 47.50 4,400 1.1%
Switzerland 2.05 102.50 6,600 1.6%
Australia 1.20 60.00 3,500 1.7%
Japan 1.15 57.50 2,300 2.5%
United Kingdom 1.85 92.50 3,100 3.0%
Germany 1.95 97.50 3,150 3.1%
China 1.15 57.50 1,100 5.2%
Brazil 1.15 57.50 600 9.6%
South Africa 1.25 62.50 550 11.4%
India 1.20 60.00 400 15.0%
Pakistan 1.00 50.00 160 31.3%
Nigeria 0.75 37.50 110 34.1%
Madagascar 1.30 65.00 105 61.9%

Its three takeaways: the "illusion of cheap gas" (a Nigerian at $0.75/L faces ~30× the strain of an American at $0.95/L); inflation multipliers where 50L exceeds 20% of pay; and subsidy traps (Nigeria pre-reform, Egypt, Sri Lanka) where decoupling from world crude ends in fiscal deficits and devaluation.

1.3 User's follow-up — the actual question, and the lived observations

"Compare to bd and recently the gov increased the fuel price from 145 to 165. The price was 125 a year ago too! Massive spike out of nowhere and gov employees get wage increase (50-200%) but private sector gets nothing. Autorikshas flood the road loadshedding spike too."

Four distinct claims to test — this drove the whole search plan:

  1. Price path 125 → 145 → 165 within about a year
  2. Government wage increase of 50–200%, nothing for the private sector
  3. Autorickshaws flooding the roads
  4. Load shedding spiking

1.4 Prior assistant's second response — tier placement

Introduced a second benchmark, the work-time metric (hours/minutes of labour per litre), and placed Bangladesh at ~31% of average monthly income per 50L tank, ranking it among the ~20–25 most fuel-stressed of 130+ tracked nations, but "not the single worst." Its tier table [UNVERIFIED]:

Tier % of monthly income (50L) Work time per 1L Representative countries
Top (hyper-affordable) 0.5–2.0% 1–4 min US, Qatar, UAE, Switzerland, Australia
Upper middle 2.0–5.0% 5–9 min UK, Germany, France, Japan, South Korea
Middle stress 5.0–15.0% 10–25 min China, Brazil, Mexico, India
Severe stress 20.0–35.0% 40–60 min Bangladesh (~31%), Pakistan (~31%), Nigeria (~34%), Sri Lanka
Extreme fuel poverty 50% to >100% 90–200+ min Madagascar, Malawi, Rwanda, Zimbabwe, Mozambique

Also argued: landlocked transport premiums (Malawi fuel $3.00–3.80/L on sub-$100 wages) and post-subsidy shock plus currency collapse (Nigeria) explain the countries worse than Bangladesh. And for Bangladesh specifically — regional outlier in South Asia at roughly double India's ~15%; private four-wheeler use restricted to the top income decile; secondary transmission through diesel freight, logistics and backup generation into food prices and transit fares.

1.5 The user's final instruction

"FROM THIS FIND OUT HOW TO RECENT SPIKE TO 145 then 165 taka for fuel in bd is gonna effect BD."

Judgment made at the start of this pass: the inherited affordability numbers are estimates of unstated provenance, and the Bangladesh figure in particular rests on a wage denominator that was never disclosed. Pass 1 therefore verifies the event first, then rebuilds the affordability arithmetic with explicit, cited denominators, then works the transmission channels. Every inherited number is treated as a hypothesis, not an input.


Part 2 — Search log

14 web searches and 5 targeted page fetches. Queries recorded verbatim.

S1. Bangladesh fuel price increase 165 taka octane petrol 2026

Confirmed the event. All four products +Tk 20/L. Octane 145→165, petrol 140→160, kerosene 135→155, diesel 115→135 (+17.4%). Effective midnight 21 September 2026. Reasons given: international prices more than doubled since March 2026; freight up on regional instability; BPC losses Tk 228.76bn (US$1.9bn) March–August; hike expected to cut annual losses ~Tk 100bn, conserve FX, curb smuggling to neighbours. Follows hikes in April and June.

S2. Bangladesh BPC fuel price September 2026 diesel octane petrol per litre

Cross-confirmed. Prices held at diesel 115 / octane 145 / petrol 140 / kerosene 135 from 1–20 September, then stepped on the 21st. Gave the clean before/after table.

S3. Bangladesh fuel price hike April June 2026 diesel octane price history timeline

Built the price path. April 18, 2026: diesel +15 → 115, octane +20 → 140, petrol +19 → 135, kerosene +18 → 130 (called a record high at the time). June 2026: all except diesel +5 → octane 145, petrol 140, kerosene 135. Confirmed the automatic pricing mechanism introduced 2024, adjusting periodically on international trends, exchange rate and import cost. This search is what let the pre-April baseline be back-computed — see D1.

S4. Bangladesh inflation rate 2026 food inflation BBS August 2026

August 2026 (BBS): general 8.26% (second consecutive monthly easing, from 8.32% in July); food 7.02% (from 7.16%); non-food 9.32% (up from 9.28%). Month-on-month general index +2.31%, food +3.86%.

S5. Bangladesh government pay scale increase 2026 salary hike public employees

Largest civil service pay rise since independence. Grades 1–10 basic up to +100%; grades 11–20 up to +142%. Lowest basic Tk 8,250 → 20,000; highest Tk 78,000 → 1,56,000. 24 lakh employees + 9 lakh pensioners; first award in 11 years. Cost Tk 1,05,580 crore. Phased over three years — 50% of the basic increase in year one, remainder in year two, house rent and allowances in year three. New: Tk 3,000/month disabled-child allowance; mobile allowance extended to all 20 grades. Confirms user claim #2 on the government side.

S6. Bangladesh loadshedding 2026 power crisis furnace oil diesel electricity generation gas shortage

Confirms user claim #4. Most serious electricity crisis since the government took office in February 2026. Gas supply at a 16-year low in July–August 2026; daily load shedding exceeded 3,500 MW at peak. Grid needs ~3,800 mmcfd, got about half. Gas plants: 12,200 MW capacity producing max 5,200 MW. PDB covering the gap with ~3,000 MW of furnace oil plus diesel peakers at peak hours. 71 of 143 power plants idle or well below capacity. Rural consumers 7–8 hrs of cuts daily, some up to 10; rural co-ops serving ~40 million people absorbed ~3,000 MW of the cuts while Dhaka fared better. Causes: Moheshkhali LNG terminal fire (21 July), rough seas idling the second floater, declining domestic gas, and QatarEnergy force majeure after Hormuz strikes — Qatar had supplied ~60% of LNG.

S7. Brent crude oil price September 2026 Hormuz war oil shock price per barrel

The upstream cause. Brent $92.30 (1 Sep) → $99.05 → $101.21 (highest close since 22 May) → $103.87 (18 Sep). Hormuz throughput collapsed 21.6 → 4.9 mbd between Q4 2025 and Q2 2026, −77%. East-West Crude Oil Pipeline shutdown 11 September. Goldman: Brent could exceed $120 in 2027 if Gulf output stays 4 mbd below prewar.

S8. Bangladesh taka exchange rate dollar 2026 foreign exchange reserves import bill oil

USD/BDT 123.03 (8 Sep 2026); 2026 average 122.69, range 121.10–124.02; taka −1.09% over 12 months. Reserves $29.47bn (IMF BPM6 method); gross reserves $35.1bn in February. Bangladesh Bank bought $4.3bn from the interbank market this fiscal year.

S9. Bangladesh bus fare increase 2026 transport fare hike after fuel price CNG

The April precedent, which became the pass-through benchmark. Bus fares +11 paisa/km: Dhaka/Chattogram metro Tk 2.45 → 2.56; 52-seat inter-district Tk 2.12 → 2.23. Owners demanded 28 paisa, citing maintenance and the dollar; government limited it to 11 "considering public interest." Diesel buses only — CNG fares unchanged. Fares to adjust automatically with fuel prices in future. Launch fares raised up to 18 paisa/km. Separately: "bus fares surge unchecked" after the hike, and a 2022 precedent of fares raised as much as 22% after a record hike.

S10. Bangladesh battery run rickshaw autorickshaw 2026 flood roads unemployment Dhaka

Confirms user claim #3. 97% of battery rickshaws and ~86% of pedal rickshaws unregistered. Innovision survey: 21.35% of battery rickshaw drivers were previously farmers — the largest occupational group. Battery cost may force fleet reductions and significant unemployment. Government considering relocating them outside metropolitan areas to fix traffic "without creating new unemployment."

S11. battery run rickshaw electricity consumption Bangladesh megawatt charging load estimate crore units

The grid-load link. Estimates vary by source: ~500 MW daily for ~60 lakh vehicles; CPD: e-three-wheelers ≈ 5% of total national generation, ~750 MW; another range 800 MW–1 GW ≈ 5% of generation. Per vehicle 4–9 kWh/day. In Dhaka: 3,300 authorised charging stations versus ~48,136 illegal charging points. August 2026: government mandated solar charging for millions of electric rickshaws.

S12. Bangladesh average monthly income labour force survey BBS per capita income 2026 median wage taka

The denominators the inherited 31% lacked. Average gross monthly salary Tk 18,000 (~$149); median monthly income Tk 9,000 (~$75); minimum wage Tk 12,500 (since 1 Dec 2023). GDP per capita $2,960 nominal / $10,850 PPP (2026 est.). Labour force 77.36 million (2024).

S13. Bangladesh furnace oil price 2026 power generation cost per unit diesel peaking plant taka kWh

The fuel nobody was discussing. Furnace oil Tk 70.10 → 94.69 (April, +Tk 24) → 113.54 (May). Private oil-fired plants received capacity payments of BDT 9.5/kWh in FY2024-25. Average generation cost rose BDT 6.61 → 11.33/kWh between FY2020-21 and FY2022-23. Bangladesh runs 10.7% of generation on oil versus India 0.02%, Pakistan 0.6%, Vietnam 0.06%. Government spending ~Tk 1.2bn/day on fuel-oil generation.

S14. Bangladesh private sector wage growth 2026 real wages minimum wage garment workers inflation

The other half of user claim #2. RMG minimum wage Tk 12,500, +56% from Tk 8,000, set December 2023 after protests; 9% annual increment. No fixed increment system in the private sector — employer discretion. National wage growth 8.05% against 8.26% inflation → real wages negative. Garment wages "barely increased since 2019" in real terms. Minimum wage adequate rurally, short of urban (especially Dhaka) cost of living.

S15. Bangladesh fuel price hike protest reaction September 2026 opposition Tarique Rahman government criticism

Political constraint. PM Tarique Rahman publicly apologised over the electricity and fuel crisis on 5 September and warned against anarchy; said the government is open to opposition input. Second fuel hike since the BNP government took office in February 2026. Energy division statement: "the ongoing war in the Middle East has caused a significant increase in the international prices of all types of petroleum products and freight charges since March 2026, and the upward trend continues." Global context: fuel-price protests across multiple countries mid-September (CNN, 15 Sep). Government to be "tougher against fuel hoarding, smuggling."

S16. Bangladesh GDP growth forecast 2026 revised 3.9% job losses remittances Gulf crisis returning migrants

Macro damage. FY2025-26 GDP growth cut from 4.6% to 3.9% on the Middle East conflict plus domestic fragilities. Gulf states ≈ 3.0% of GDP in remittances ≈ 50% of total inflows. Poverty exits cut from 1.7 million to 0.5 million in FY2026. Real wages already negative for low-income and unskilled workers across agriculture, industry and services.

S17. Bangladesh diesel irrigation boro season cost farmers 2026 agriculture subsidy

The largest forward-looking channel. Boro irrigation Tk 3,500–4,000/bigha, hitting Tk 6,000 during fuel shortages in Chuadanga. ~15 lakh shallow tube wells run on diesel. Diesel reportedly sold Tk 15–80/L above regulated price in some regions. Boro is >55% of total rice production, ~21–22 million tonnes. Existing mechanisation subsidies (70% haor / 50% elsewhere) cover combine harvesters, not diesel. Experts calling for direct cash support.

S18. Bangladesh garment RMG factories diesel generator cost 2026 export competitiveness energy crisis production loss

Export channel. Factories paying a Tk 10/L premium for generator diesel; some burning up to $25,000/day (Tk 30 lakh) on alternative fuels; smaller units ~Tk 40,000/day with 3–4 hours of outages. Production capacity down 20–30%; knit composite and woven washing plants at 50% or less. Risk of losing orders to India, Pakistan, Vietnam.

S19. Bangladesh LDC graduation November 2026 duty free access RMG exports impact

The collision date. Graduation 24 November 2026. LDC status currently gives duty-free access to EU, UK, Canada, Japan covering 75–78% of exports; RMG is >80% of export earnings and 4m+ workers. Post-transition tariffs 9–12% EU, 7–13% Japan, 16–18% Canada; estimated 6–14% drop in export earnings, $1–7bn/yr (WTO warning of up to $8bn ≈ 14% of exports). EU/UK/Canada grace periods to November 2029.

S20. Bangladesh BPC losses subsidy 2026 fuel smuggling India automatic pricing formula next adjustment

BPC mechanics. Losses Tk 22,875.66 crore March–August 2026, driven by global prices rising without domestic adjustment — the automatic formula exists but was not applied after March. India's prices are deliberately built into the formula to deter cross-border smuggling. Formula components: import cost at the payment-day exchange rate, plus customs tax, dealer commission, BPC development cost, operating expense, transport, and a small margin; monthly adjustment. Precedent exists for cuts (Tk 2/L reduction under the formula).

Fetch F1 — Daily Star, "Fuel prices hiked by Tk 20 a litre"

Prices, effective date, April comparison. Notably contained no rationale, BPC figures, impact estimate, official quotes, or mention of transport/irrigation/power effects — pure price reporting. Recorded because the absence is itself informative about the announcement's framing.

Fetch F2 — Daily Star, "Fuel price hike lays bare stress on economy" (April)

The April analogue, used to calibrate expectations. Energy Minister Iqbal Hassan Mahmood Tuku: "Fuel must be purchased with foreign currency. By increasing prices slightly, we are trying to keep the situation at a 'tolerable level.'" Transport associations proposed city fare rises up to 64% and long-distance up to Tk 3.80/km; waterways proposed 36–42%. Agricultural economist Jahangir Alam Khan: severe Boro risk at peak irrigation demand, threatening yields across irrigation, harvesting, transport and marketing. Prof Mustafizur Rahman (CPD): will "erode purchasing power after prolonged inflation, while raising costs across transport, industry, and trade." Zahid Hussain (ex-World Bank) supported the adjustment — subsidies "drained public funds and disproportionately benefited wealthier groups." BPC raised supply targets from 20 April: diesel +17.5%, octane +25.9%, petrol +20.6%.

Fetch F3 — Daily Star, "BPC may run out of funds for fuel imports in Oct" ⚠️ the decisive find

Fetch F4 — TBS, "Fuel prices hiked by Tk20 per litre"

Percentages: diesel 17.4%, petrol 14.3%, octane 13.8%, kerosene 14.8%. Government kept prices through August "despite doubling international costs." Rationale: reduce BPC losses, prevent smuggling, maintain electricity and gas subsidies. BPC loss Tk 89/L on diesel; ~Tk 109 crore/day; ~Tk 40,000 crore/year annualised; hike saves ~Tk 10,000 crore/year. Regional diesel comparison: Bangladesh 135, India 134.76, Myanmar 164.83, Nepal 161.24, Pakistan 185.48. Reactions — Dr Fahmida Khatun (CPD): "wide-ranging consequences," transport costs into food prices, disproportionate burden on the poor. Anwar-ul Alam Chowdhury Parvez (BCI): "more than 15% increase without consulting anyone," intensified inflation, reduced export competitiveness. Shams Mahmud (Bangladesh-Thai Chamber): necessary buffering but adds inflationary pressure on already-weakening export industries. No statement on further hikes, rationing or supply restrictions.

Fetch F5 — CPD, Fahmida Khatun on the hike

"Substantial policy intervention, especially when inflation and household financial pressure are already high"; may "intensify inflation, reduce household consumption and weaken growth." Lower-income families disproportionately burdened. Irrigation: 70% of pumps diesel. Manufacturing hit on generators and logistics; small businesses most vulnerable. Recommendations: disclose the price calculation transparently, prevent excessive fare rises and market manipulation, targeted support for low-income households, small farmers and public transport users. No quantified projections offered — which is why Part 4 builds its own.

Fetch F6 — Daily Star editorial, "Hurrah for some, pain for many"

The distributional case. 24 lakh employees and 9 lakh pensioners gain after an 11-year freeze; the general population bears inflation and reduced public services because the government lacks revenue and must borrow. Implementation cost Tk 1,05,580 crore. Last year's tax revenue shortfall Tk 88,000 crore against a Tk 5,03,000 crore target. Projected job losses from regional instability: 6,00,000. Growth forecast cut 4.5% → 3.9%. Contains no private-sector wage data — only a suggestion that the private sector "align its pay structures without putting additional strain on the economy." Noted as a gap the log fills from S14.

Fetch F7 — search on mitigation measures

Family Card: digital social safety-net programme launched by the Ministry of Social Welfare 10 March 2026, unifying poor and low-income families in one database for direct financial and food assistance. Cash transfers/vouchers suggested as the effective instrument. No TCB/OMS measures found tied to this specific hike. World Bank has a fuel-market-volatility support factsheet (May 2026).


Part 3 — Consolidated data table

3.1 Price path, Tk/litre [REPORTED] except D1

Fuel Pre-18 Apr [DERIVED D1] 18 Apr June 21 Sep Δ since March [DERIVED] Δ this hike
Diesel 100 115 115 135 +35.0% +17.4%
Petrol 116 135 140 160 +37.9% +14.3%
Octane 120 140 145 165 +37.5% +13.8%
Kerosene 112 130 135 155 +38.4% +14.8%
Furnace oil 70.10 94.69 113.54 (May) 113.54 +62.0% n/a

3.2 BPC position

Item Value Source
Losses, March–August 2026 Tk 22,875.66 crore F3, F4, S1
Loss per litre, diesel, pre-hike Tk 89 F4
Daily diesel loss, pre-hike Tk 109 crore F4
Annualised diesel loss run-rate ~Tk 40,000 crore F4
Claimed annual saving from hike ~Tk 10,000 crore F4
Working capital, 6 Sep 2026 Tk 12,368 crore F3
Stated minimum working capital Tk 15,000–20,000 crore F3
Drawn from development accounts Tk 19,500 crore (Tk 1,679.72 crore left) F3
August import spend Tk 7,603 crore (vs Tk 2,741 crore YoY, +177%) F3
Duties + taxes on diesel, August Tk 32.44/litre F3
BPC proposed September diesel price Tk 187 (government set 135) F3
LC funding exhaustion risk October 2026 F3

3.3 Macro

Indicator Value Source
Inflation, general, Aug 2026 8.26% (2nd month of easing) S4
Inflation, food / non-food 7.02% / 9.32% S4
MoM general index +2.31% S4
National wage growth 8.05% (below inflation) S14
USD/BDT 123.03 S8
Reserves (BPM6) $29.47bn S8
GDP growth FY26, revised 3.9% (from 4.5–4.6%) S16, F6
Projected job losses 600,000 F6
Poverty exits FY26 0.5m (from 1.7m pre-conflict) S16
Gulf share of remittances ~50% of inflows, ~3% of GDP S16
Brent, September 2026 $92–104 S7
Hormuz throughput 4.9 mbd (from 21.6, −77%) S7

3.4 Wages

Group Level Change Source
Govt grades 1–10 — basic up to +100% S5
Govt grades 11–20 lowest 8,250 → 20,000 up to +142% S5
Govt highest basic 78,000 → 1,56,000 +100% S5
Covered population 24 lakh + 9 lakh pensioners first rise in 11 years S5
Cost / phasing Tk 1,05,580 crore over 3 years 50% of basic in year 1 S5
RMG minimum wage Tk 12,500 (since Dec 2023) +9%/yr increment only S14
Private sector generally no mechanism employer discretion S14
Average gross salary Tk 18,000 (~$149) — S12
Median monthly income Tk 9,000 (~$75) — S12

3.5 Power and e-rickshaws

Item Value Source
Peak load shedding, Jul–Aug 2026 >3,500 MW S6
Gas supply 16-year low; ~half of 3,800 mmcfd need S6
Gas plants 12,200 MW capacity → max 5,200 MW output S6
Idle/derated plants 71 of 143 S6
Furnace-oil generation ~3,000 MW; ~Tk 1.2bn/day S6, S13
Oil share of generation 10.7% (India 0.02%, Pakistan 0.6%) S13
Rural outages 7–8 hrs/day, up to 10 S6
e-3W grid load 500 MW–1 GW ≈ 5% of generation S11
e-3W per-vehicle draw 4–9 kWh/day S11
Dhaka charging points 3,300 authorised vs ~48,136 illegal S11
Unregistered battery rickshaws 97% S10
Drivers who were farmers 21.35% S10

3.6 Agriculture and industry

Item Value Source
Diesel share of irrigation pumps ~70% F5
Diesel shallow tube wells ~15 lakh S17
Boro share of rice output >55%, 21–22 mn tonnes S17
Irrigation cost/bigha Tk 3,500–4,000 (to 6,000 in shortages) S17
Rural diesel over-pricing Tk 15–80/L above regulated S17
Boro procurement price held at old level S17
PDB dues sought pre-Boro Tk 9,750 crore S17
RMG output loss 20–30% (washing plants ≤50%) S18
Generator diesel premium Tk 10/L over pump S18
Factory fuel spend Tk 40,000/day small; up to $25,000/day large S18
LDC graduation 24 Nov 2026; grace to Nov 2029 S19
Post-graduation tariffs EU 9–12%, Japan 7–13%, Canada 16–18% S19
Export loss estimate 6–14% ($1–8bn/yr) S19

Part 4 — Derived calculations

Every number in impact_analysis.md marked est. is defined here.

D1 — Pre-April baseline prices. April hike deltas subtracted from April levels: diesel 115−15=100; petrol 135−19=116; octane 140−20=120; kerosene 130−18=112. → Cumulative since March: diesel +35.0%, petrol +37.9%, octane +37.5%, kerosene +38.4%. Note: the user recalled octane at 125 a year ago; 120 is the immediate pre-April level. Both are consistent with a formula drifting in the 120–131 band through 2025. Octane 125 → 165 = +32% year-on-year.

D2 — 50L octane affordability. 50 × 165 = Tk 8,250 = $67.07 at 123.0 BDT/USD.

D3 — Work-time per litre. At 208 hrs/month: avg wage Tk 86.54/hr → 165 ÷ 86.54 = 1.91 hrs = 1h 54m. Median Tk 43.27/hr → 3.81 hrs = 3h 49m. RMG Tk 60.10/hr → 2h 45m. Diesel at avg wage: 135 ÷ 86.54 = 1.56 hrs = 94 min. → Places Bangladesh in the inherited table's Extreme Fuel Poverty band (90–200+ min/L), not Severe Stress (40–60 min).

D4 — Diesel gap closure. BPC formula price 187 − pre-hike 115 = Tk 72 gap; Tk 20 passed = 27.8% closed; Tk 52 remaining. Note the flat Tk 20 over-recovers on petrol (160 vs 150 proposed), octane (165 vs 154) and kerosene (155 vs 146) — a cross-subsidy from those three to diesel that still leaves diesel short.

D5 — Diesel volume and saving cross-check. Implied throughput = 109 crore/day ÷ Tk 89/L = ~12.2 mn L/day. Saving = 20 × 12.24m × 365 = ~Tk 8,900 crore/yr against the government's claimed ~Tk 10,000 crore. ✓ Internally consistent, which corroborates the reported loss figures. Post-hike loss = 89 − 20 = Tk 69/L; daily ~Tk 85 crore; annualised ~Tk 31,000 crore. Monthly cash relief ~Tk 730 crore against a ~Tk 3,800 crore/month loss run-rate (22,876 ÷ 6).

D6 — The duty wedge. Tk 32.44 of the Tk 135 pump price = 24.0%. Net public-sector position per litre: NBR collects 32.44, BPC loses 69 → net −Tk 36.56/L. Zeroing the duty would close 32.44 ÷ 52 = 62.4% of the remaining gap with no pump-price rise.

D7 — Inflation decomposition [ASSUMED] weights, standard for the Bangladesh CPI basket:

Channel Assumption Contribution
Direct fuel purchase 1.5–2.5% weight × ~15% price rise +0.25 to +0.40 pp
Transport services 5–6% weight × 8–15% realised fare rise +0.40 to +0.90 pp
Food via freight + irrigation 46% food weight × 0.8–1.4% cost pass-through +0.40 to +0.65 pp
Cost-push subtotal +1.05 to +1.95 pp
Pay-scale demand injection ~Tk 30–35k crore ≈ 0.55% of GDP +0.30 to +0.50 pp
→ From 8.26%, expect 9.5–10.5% by Dec 2026–Jan 2027, non-food leading to ~10.5–11%, food re-accelerating on a lag. The two-month easing trend breaks in the October or November print.

D8 — GDP denominator for D7. $2,960 per capita × ~175m = ~$518bn ≈ Tk 63.7 lakh crore. Year-one pay injection Tk 30–35k crore = 0.47–0.55% of GDP.

D9 — Bus fare pass-through. April: diesel +15.0% → gazetted metro fare +4.49% (2.45→2.56). Ratio 0.30. Applied to +17.4%: +5.2% → metro Tk 2.69/km, inter-district Tk 2.35/km. Commuter effect, 20 km/day round trip: gazetted +Tk 2.6/day ≈ Tk 60/month; but a minimum-fare round-up from Tk 10 to 15 on two legs × 22 working days = Tk 220–300/month ≈ 1.8–2.4% of an RMG wage. Realised overshoot is the documented pattern (S9: "surge unchecked"; 2022: up to 22%).

D10 — Irrigation cost. Diesel ≈ 60–70% of the Tk 3,500–4,000/bigha baseline ≈ Tk 2,400. This hike: +17.4% × 2,400 = +Tk 418/bigha ≈ +Tk 1,254/acre (3 bigha/acre). Since March: +35% × 2,400 = +Tk 840/bigha ≈ +Tk 2,520/acre. Boro area ~4.9m ha = 12.1m acres; ~70% diesel-irrigated = 8.47m acres. Aggregate: ~Tk 1,060 crore from this hike; ~Tk 2,130 crore since March. Per kg of rice: diesel-irrigation ≈ Tk 3–4 of Tk 32–35/kg production cost; +35% = +Tk 1.05–1.40/kg.

D11 — Generator costs. Effective factory diesel = 135 + 10 premium = Tk 145/L. A unit reported at Tk 40,000/day scales to ~Tk 47,000/day (+17.4%); one at $25,000/day → ~$29,300/day.

D12 — Freight. 15t truck, Dhaka–Chattogram 250 km at ~3 km/L = 83.3 L one way × Tk 20 = +Tk 1,667/trip ≈ 6% of a Tk 25,000–30,000 charter. Realised rate rise 8–12% once empty return legs and driver costs compound.

D13 — e-3W versus liquid fuel, per km. Petrol 3W at Tk 165/L and ~15 km/L = Tk 11.00/km. Battery 3W drawing 4–9 kWh/day over ~100 km at Tk 8–9/kWh = 0.04–0.09 kWh/km = Tk 0.32–0.81/km. Ratio ~14–34×; for the 93% of Dhaka charging points that are illegal, the operator's marginal cost is ~0. The hike widened an already decisive wedge by another 13.8%.

D14 — Forex. August import spend Tk 7,603 crore ÷ 123 = $618m/month vs Tk 2,741 crore = $223m a year earlier. Increment $395m/month ≈ $4.74bn/yr ≈ 16.1% of the $29.47bn reserve stock. Demand destruction at short-run price elasticity −0.1 to −0.2 [ASSUMED] → volume −1.7 to −3.5% → ~$150–250m/yr saved. Real but second-order against a $4.74bn increment.

D15 — Smuggling arbitrage. Pre-hike: India 134.76 − Bangladesh 115 = Tk 19.76/L incentive to move subsidised diesel out. Post-hike: 134.76 − 135 = −Tk 0.24/L. Arbitrage eliminated. This rationale is sound.

D16 — Wage wedge. Year-one phased grade-20 basic = 8,250 + 50% × (20,000 − 8,250) = Tk 14,125 (+71%, reaching +142% by year two). RMG worker: Tk 12,500, +0% since Dec 2023 (9%/yr increment aside). Against ~30% cumulative inflation since 2023 [ASSUMED], the first is real-positive and the second sharply real-negative. Covered population 33 lakh ≈ 2% of population, ~4% of the 77.4m labour force.


Part 5 — Findings against the four original claims

# User's claim Verdict Evidence
1 "145 to 165, was 125 a year ago" Confirmed Octane 145→165 on 21 Sep; ~120–125 a year ago. +32% YoY, +37.5% since March
2 "Govt +50–200%, private nothing" Confirmed, magnitude slightly overstated at the top Grades 1–10 up to +100%, grades 11–20 up to +142% (not 200%); private sector has no mechanism, real wages negative (8.05% growth vs 8.26% inflation)
3 "Autorickshaws flood the road" Confirmed, and causally linked Operating cost 14–34× cheaper per km (D13); 97% unregistered; fleet draws 5% of national generation
4 "Load shedding spike" Confirmed, and part of the same loop >3,500 MW peak cuts; gas at 16-year low; e-3W charging 500 MW–1 GW on 48,136 illegal Dhaka points
— "Out of nowhere" Partly rejected The shock is external (Hormuz) and was flagged since March. What was "sudden" is the policy: the monthly formula was suspended March–September, then six months of adjustment released in one Tk 20 step

The finding that mattered most, and was not in the originating conversation at all: BPC asked for Tk 187 diesel and got Tk 135, leaving Tk 52/L unpassed with only 27.8% of the gap closed, while its working capital covers September only and LC funding may be exhausted in October. Tk 20 is roughly a quarter of the adjustment, not the end of it.


Part 6 — Open questions for Pass 2

  1. Actual BBS CPI basket weights for fuel, transport and food — D7 uses assumed weights and is the least firm calculation in the set.
  2. Diesel volume split: irrigation vs freight vs generators vs power. Would sharpen D10/D11/D12 considerably.
  3. Whether the diesel duty cut (D6) is under active consideration — no reporting found either way.
  4. Gazetted bus fare revision after this hike — not yet published at the time of this pass; D9 is a forecast to be scored.
  5. Boro acreage intentions for 2026-27 — the single most important unknown for 2027 food inflation.
  6. Real-wage series for informal workers specifically, not the national aggregate.
  7. Whether the Family Card rail is actually being used for fuel compensation.
  8. e-3W fleet size: estimates range from 500,000 to 6,000,000 across sources. The load estimates rest on this and the range is unacceptably wide.

Sources

The hike

Price history

BPC and pricing formula

Economy and inflation

Transport fares

Agriculture

Power and load shedding

Battery rickshaws

RMG and exports

Wages and pay scale

Global oil context

Politics and mitigation