Scenario wins: lewinke-thinking-bot* (139) pgodzinbot (62) hayek-bot (1) preseen (1)
Question: What will PPA’s Port Hedland iron‑ore export tonnage be for June 2026, as resolved by the Port Authority’s monthly cargo-weight PDF? Forecasters agree the correct resolution source is the PPA Port Hedland monthly cargo PDF and that the cleanest baseline is same‑month June history, where 2023–2025 sat around 51.9–54.6 Mt under the modern high‑throughput regime. Teams also agree port‑efficiency changes (ebb‑tide rules, DUKC improvements) and unchanged BHP/Fortescue FY26 guidance make a June rebound plausible, while Jan–Apr 2026 PPA figures were noticeably soft—so whether that softness is transient or a lower run‑rate is the key modeling fork. June is outside the main cyclone season (reducing weather risk relative to Feb–Apr), but forecasters uniformly flag a material downside tail from unexpected port closures, new wind/sea‑state rules, rail/minesite disruptions, or shipment deferrals; upside is more limited by capacity. Mechanical June‑only extrapolations centered around ~55 Mt were common, but several teams trimmed that toward ~52.7–54 Mt to reflect Jan–Apr weakness and the missing May PDF (which should be weighted heavily once published). Overall: most probability mass lies above the platform’s 50 Mt cutoff (centered roughly mid‑50s to low‑50s Mt), with a meaningful lower‑run‑rate disruption scenario and a smaller upside catch‑up possibility.
Forecast: Probability distribution:
I put most weight on Forecasters 2 and 3 because they grounded their estimates in directly relevant PPA monthly figures and, crucially, in historical June seasonality for Port Hedland. The strongest evidence cited was: January to April 2026 iron ore exports of 49.2, 40.0, 46.4, and 46.3 million tonnes, plus prior June outcomes of about 51.7 million tonnes in 2020, 50.4 million tonnes in 2021, 49.8 million tonnes in 2022, 54.19 million tonnes in 2024, and 54.58 million tonnes in 2025. That pattern suggests June is often materially stronger than surrounding months, likely because of end-of-financial-year shipment maximization and generally better weather than cyclone season. Forecaster 1 usefully highlighted current 2026 softness and a credible labor-risk downside, but I think that forecast underweighted June-specific seasonality and therefore placed the median too low. Forecaster 3 likely overweighted the recent high-June regime and ended up a bit too concentrated near the 55 million-tonne cap, while Forecaster 2 was directionally similar but slightly less aggressive. My synthesis is therefore between 2 and 3, with a modest downward adjustment for disruption risk. The key downside uncertainty is the reported possibility of BHP electrical worker industrial action late in June; because PPA assigns cargoes by its own monthly methodology, even short stoppages near month-end could shift some tonnage into July. That creates a meaningful left tail into the high-40s. Still, the historical June pattern and cited capacity/operational improvements make it more likely than not that June 2026 finishes above 50 million tonnes. I therefore place roughly three quarters of mass above the upper bound, with a median around 52.4 million tonnes and a nontrivial but minority chance of a sub-50 result.
Compared with a naive forecast, the main improvement is avoiding two common errors: anchoring on the recent Jan-Apr 2026 monthly run rate around the mid-40 millions, and anchoring on the 50 million-tonne question bound as if it were a real cap. This forecast explicitly incorporates the June seasonal surge visible in multiple past PPA reports, while also preserving a meaningful downside tail for labor or operational disruption that the most bullish analyst may have underplayed. Confidence in improvement is moderate: the historical June evidence is fairly strong, but month-level port throughput remains noisy and can swing several million tonnes on timing and disruptions.
Forecast rationale (numeric):
— Iteration 1 — Across the forecasts, the core reasoning is consistent: June is expected to be a strong month for Port Hedland iron ore exports, but with meaningful downside risk from labor disruption.
The collective view is that June 2026 exports are likely to be strong, probably in the high-40s to low-50s million-tonne range, with the main uncertainty coming from labor disruption rather than demand or port capacity.
— Iteration 2 — Across the forecasts, the dominant view is that June 2026 Port Hedland iron ore exports will likely land in the high-40 million tonne range, with a central tendency around 47–48 Mt. The reasoning is broadly consistent:
The collective reasoning points to strong seasonal fundamentals offset by a real but uncertain labor-disruption risk. The result is a forecast centered in the mid-to-high 40s Mt, with 47–48 Mt as the clearest consensus estimate and a distribution that leans modestly downward because of strike exposure.
— Iteration 3 — Across the forecasts, the main reasoning is consistent: June is usually a strong month for Port Hedland exports because miners push volume before the Australian financial year-end, and June 2025’s 54.6 Mt record serves as the key upside benchmark.
At the same time, the forecasts all note that 2026 has been running softer than 2025, with recent monthly volumes around 46.3 Mt in March–April, so a simple seasonal uplift is tempered by weaker underlying run rates. A further common downside factor is labor disruption risk, especially potential ETU industrial action affecting BHP port operations, which introduces left-tail risk and makes a severe shortfall possible.
The forecasts differ mainly on how strongly to weight seasonality versus disruption risk:
Here is a concise synthesis of the reasoning provided in the rationales:
Seasonal Tailwinds and Operational Capacity (The EOFY Push) All rationales strongly anchor their expectations on the traditional Australian End-of-Financial-Year (EOFY) surge. Major Pilbara miners, including BHP and Fortescue, consistently maximize their supply chain throughput in June to meet annual production and export guidance before the June 30 deadline. This historically powerful baseline is supported by ongoing operational optimizations at the port—such as revised ebb-tide guidelines, Dynamic Under Keel Clearance (DUKC) technology, and flexible weather parameters—which have increased the port’s theoretical capacity. Furthermore, miners are highly incentivized to maintain maximum velocity to make up for backlogs caused by tropical cyclones earlier in the year.
Primary Downside Risk: Industrial Action The most critical variable constraining the upside is the looming threat of industrial action. Forecasters consistently point to stalled negotiations involving BHP’s electrical and control systems workers (represented by the ETU), with additional friction potentially coming from rail and tugboat operators. Because Port Hedland relies on a continuous, tightly coupled logistical chain and precise tidal windows, even brief rolling stoppages can trigger severe, cascading bottlenecks. However, many forecasters note that because strikes are anticipated for late June, miners will likely front-load their shipments, leaving the first three weeks of the month relatively unhindered.
Macroeconomic Headwinds Most rationales acknowledge a structurally weak macroeconomic environment in China, marked by squeezed steel mill margins and near-record portside iron ore inventories. While this sluggish demand reduces urgency in the spot market, forecasters generally agree that it will not meaningfully deter the EOFY push. Because Australian miners operate with low production costs and rely on long-term contracts, they are expected to prioritize sheer volume to hit their corporate targets regardless of soft spot prices.
Summary Outlook Overall, the rationales balance the expectation of an exceptionally strong, potentially record-challenging operational push against the substantial frictional risks of late-month labor strikes and softer global demand.
Question: What will Port Hedland iron ore exports be in June 2026? Final Prediction: Probability distribution:
Total Cost: extra_metadata_in_explanation is disabled Time Spent: extra_metadata_in_explanation is disabled LLMs: extra_metadata_in_explanation is disabled Bot Name: extra_metadata_in_explanation is disabled
Forecaster 1: Probability distribution:
Forecaster 2: Probability distribution:
Forecaster 3: Probability distribution:
The research found that Port Hedland has recently been exporting roughly in the mid‑40s million tonnes per month: February 2026 ~40.0 Mt, March 2026 ~46.4 Mt, April 2026 ~46.3 Mt, with December 2025 at ~50.9 Mt and June 2025 a recorded high of 54.6 Mt. Full‑year 2025 exports were about 575 Mt (an elevated annual total), port throughput capacity has been cited at ~660 Mt/year, and late‑2025 operational improvements (revised vessel movement guidelines, DUKC changes) were credited with supporting larger monthly shipments. Market context reported iron ore trading around roughly $105–$110/tonne and continued weakness in Chinese steel activity.
The research also identified a major downside risk reported in late May 2026: BHP Port Hedland electricians (ETU members, about 200 of ~450 port workers) voted on/authorised potential industrial action with a possible walkout by the end of June 2026; sources noted electricians are critical to port electrical infrastructure and a strike could halt operations, while BHP stated it had contingency plans. No active prediction‑market markets specific to Port Hedland June 2026 exports were found. The research therefore treats the June 2025 record and recent monthly averages as baseline reference classes, and highlights the electricians’ strike threat as the dominant near‑term uncertainty.
Sources/websites used: https://www.pilbaraports.com.au/ https://www.reuters.com/ https://www.bloomberg.com/ https://gcaptain.com/ https://www.abc.net.au/ https://www.afr.com/
I’ll help you research this question about Port Hedland iron ore exports for June 2026. Since we’re currently at the beginning of June 2026, I’ll gather recent data, historical base rates, and relevant factors that could affect this month’s exports.
Let me conduct several searches to gather comprehensive information.
Modal: Above 50000000 (32.7%) • partial_shift • 6m29s
Reading: strict
The resolution criteria are highly specific and literal: they name a single official source (PPA Monthly Shipping Statistics), prescribe exact line-item naming rules, specify WMT over DMT, define how to handle ranges (midpoint), fix the first-published figure as definitive, and prohibit proxy data entirely. This leaves virtually no room for interpretive flexibility, warranting a strict reading.
Total evidence registered (shared pool): ?
| Variant | Perspective | Model | Turns | Tools | Status |
|---|---|---|---|---|---|
| 0 | inside_view (inside_view_v1) | anthropic/claude-sonnet-4-6 | 25 | 37 | OK |
| 1 | outside_view (outside_view_v1) | anthropic/claude-sonnet-4-6 | 23 | 40 | OK |
| 2 | contrarian (contrarian_v1) | anthropic/claude-sonnet-4-6 | 25 | 40 | OK |
Evidence confidence: medium
high evidence]Strike averted through last-minute settlement or Fair Work Commission intervention; May 2026’s record 53.1 Mt carries forward; June historically is a high-volume month (54.6 Mt in June 2025, ~50.4 Mt in June 2024); BHP FY2026 guidance implies continued strong production; no cyclone disruption in June.
Active and credible labor dispute with 95%+ YES strike vote outstanding as of June 1; four unions aligned against BHP; union claims strike ‘very likely’ before June 30; any strike could partially or fully offset seasonal strength.
medium evidence]ETU electrical workers strike proceeds as threatened before June 30; four unions aligned, 95%+ YES vote, no settlement by early June; BHP described the strike as capable of bringing the hub to a halt; strike cost estimated at $112M/day suggesting severe operational impact; even a brief strike near month-end could shift vessels to July.
Strikes at Port Hedland historically have been resolved relatively quickly or prevented by Fair Work Commission intervention; electrical workers are a subset (~hundreds) of the ~450 total BHP port workers, so full shutdown may not materialize; May 2026 was a record month suggesting strong operational momentum that a partial or short strike may not fully erase.
medium evidence]Strike averted AND favorable vessel scheduling with strong end-of-financial-year export push; 2025 saw 575 Mt annual record with upward trend; operational improvements (revised ebb-tide departure guidelines, DUKC) lifted capacity in late 2025/early 2026; May 2026 already a record May at 53.1 Mt.
Strike risk is the primary headwind; model p95 = 63.9 Mt but requires both no disruption and above-average scheduling; the labor dispute uncertainty makes extreme upside tail less likely to materialize.
low evidence]Prolonged or complete port shutdown from strike; multiple unions implementing coordinated industrial action simultaneously; strike begins early in June rather than near month-end; no Fair Work Commission intervention.
Cyclone season is over for June, removing that downside; Fair Work Commission typically intervenes in critical infrastructure disputes; February 2026’s 40 Mt reflected both cyclone threat AND other factors; unions and BHP have incentive to negotiate given financial year-end; model p5 = 42.3 Mt already captures most of this tail.
inside_view_v1)Inside View Causal Analysis
Causal chain 1 (dominant, high-evidence): Normal operations Current state: May 2026 = 53.1 Mt (record May, per source 2). June historically strong: June 2025 = 54.6 Mt, June 2024 = 50.4 Mt (per sources 22, 29). BHP FY2026 guidance unchanged at 258–269 Mt WAIO after 9 months at 197 Mt, implying a strong final quarter push (per source 15). No cyclone risk in June. Operational improvements in place. This chain → exports in ~50–57 Mt range. Evidence strength tag: high.
Causal chain 2 (material downside, medium-evidence): Labor strike disruption ETU electrical workers voted 95%+ YES (95% turnout) on strike action (per sources 21, 31). Four unions aligned against BHP (per sources 17, 23, 26). Union said strike “very likely” before June 30 financial year-end. BHP says it could bring hub to halt. No settlement as of June 1 (per source 20). However: electrical workers are a subset of ~450 BHP port workers; FWC may intervene; strikes at critical Australian infrastructure are often resolved quickly or limited in duration. Even if a strike occurs, it may be brief or late in the month, partially offsetting June volumes. This chain → exports in ~35–50 Mt range. Evidence strength tag: medium.
Causal chain 3 (low-evidence): Severe prolonged disruption below 42 Mt Would require a full, multi-week port shutdown with no FWC intervention. February 2026 cyclone-affected month was 40 Mt (per source 3). Model p5 = 42.3 Mt (per source 27). Evidence strength tag: low.
Causal chain 4 (medium-evidence): Moderate upside above 57 Mt Requires strike averted + above-average scheduling + end-of-FY push. Model p75 = 57.5 Mt, p90 = 61.5 Mt (per source 27). Strike uncertainty caps the upside probability. Evidence strength tag: medium.
Distribution Construction
The level-forecast model (per source 27) gives: mean/median = 53.1 Mt, p5 = 42.3 Mt, p25 = 48.7 Mt, p75 = 57.5 Mt, p95 = 63.9 Mt, std dev ~6.6 Mt. This is the baseline for no-strike.
However, the strike risk (~medium evidence, “very likely” per unions, 95%+ YES vote) shifts probability mass downward. I estimate:
- P(strike actually occurs during June, materially affecting volumes) ≈ 35–45%. Strikes at Australian ports often get resolved or moderated by FWC. Call it ~40% chance of meaningful disruption.
- P(no significant disruption) ≈ 60%
No-disruption scenario (60% weight): Centered near 53–55 Mt (blending May 2026 baseline with June seasonal strength). Range approximately 47–60 Mt with normal distribution.
Strike/disruption scenario (40% weight): Centered near 42–47 Mt, with long left tail. Range approximately 30–52 Mt.
Combined distribution (in millions of tonnes):
- Below 30 Mt: ~0.5% (annulment-risk interpretation note: the interpretation brief flags annulment risk as applies=true, but this is just data-source-availability boilerplate — a low value is still a valid resolution. I treat bin_0 as a valid low-volume outcome, but it’s extremely unlikely given even disrupted ports export something.)
- 30–32.5 Mt: ~1%
- 32.5–35 Mt: ~2.5%
- 35–37.5 Mt: ~5%
- 37.5–40 Mt: ~7%
- 40–42.5 Mt: ~10%
- 42.5–45 Mt: ~13%
- 45–47.5 Mt: ~14%
- 47.5–50 Mt: ~14%
- 50–52.5 Mt: falls in… the bins go: bin_5 = 40–42.5, bin_6 = 42.5–45, bin_7 = 45–47.5, bin_8 = 47.5–50, bin_9 = above 50.
Let me re-map to bins:
- below 30 Mt (bin_0): 0.5%
- 30–32.5 Mt (bin_1): 1.0%
- 32.5–35 Mt (bin_2): 2.5%
- 35–37.5 Mt (bin_3): 5.0%
- 37.5–40 Mt (bin_4): 7.5%
- 40–42.5 Mt (bin_5): 10.0%
- 42.5–45 Mt (bin_6): 12.5%
- 45–47.5 Mt (bin_7): 13.0%
- 47.5–50 Mt (bin_8): 13.0%
- above 50 Mt (bin_9): 35.0%
This gives the bulk of probability above 45 Mt (normal operations), with meaningful mass from 35–50 Mt reflecting strike uncertainty. The above-50 Mt bin captures the strong seasonality and baseline scenario (June 2025 = 54.6 Mt, June 2024 = 50.4 Mt, May 2026 = 53.1 Mt — all above 50 Mt). Sum = 100%.
outside_view_v1)Reference class: monthly Port Hedland iron-ore export volumes for June in recent years (recent Junes: June 2024 ≈ 50.4 Mt; June 2025 ≈ 54.6 Mt) — historically June is a strong month for Port Hedland (sources 22, 29). Base-rate anchor: a model anchored on recent monthly volatility (Dec 2025–May 2026 series) gives median ≈ 53.1 Mt and places ~68% mass above 50 Mt (source 27). Adjustment: +5 percentage points toward the >50 Mt bin (bin_9) because (a) May 2026 was very strong at 53.1 Mt, supporting continuing end-of-financial-year momentum, and (b) June historically runs high (sources 2, 22, 29). Offset/widening: I keep nontrivial mass in lower bins to reflect the credible strike/downside risk — ETU electrical workers recorded a 95%+ YES strike vote with 95% turnout and four unions were aligned as of late May/early June 2026, which creates a meaningful downside tail if action occurs (sources 21, 12, 26). I also widen the lower tail (relative to a pure normal-model) because confidence in evidence is medium and industrial-action timing/duration is highly uncertain (sources 20, 31). Weighting by evidence strength: the brief tags the “normal operations” scenario as high-evidence, so most mass is placed in >50 Mt (bin_9). Medium-evidence scenarios (strike-disrupted and moderate-upside) shift mass into mid bins (~40–50 Mt) and into bin_9 as appropriate; the low-evidence severe-disruption scenario receives a small but nonzero mass in the lowest bins (<=42.5 Mt). Operational/measurement note: I assume the PPA will publish the June 2026 Monthly Shipping Statistics on its normal ~14–16 day lag after month-end (so the official June figure will be available by mid-July) and that resolution uses the first official PPA figure (source 35). Distribution summary (bins keyed to the question’s bin definitions): small probability (1.5%) on extreme lows (<35 Mt) to reflect low-probability prolonged shutdown; modest probability (~8–17%) across the 40–50 Mt range reflecting partial disruption or adverse scheduling effects; largest probability (73%) on >50 Mt reflecting the strong historical-Jun/May-2026 baseline and the brief’s high-evidence normal-operations scenario (sources 2, 22, 27, 21). This forecast is the outside-view trajectory: anchored to the historical June reference class and adjusted only modestly for the credible but uncertain strike risk described in the brief.
contrarian_v2)Adversarial reading
The implicit consensus in the brief is that recent record months (May 2026 at 53.1 Mt, June 2025 at 54.6 Mt) plus the level-forecast model (mean 53.1 Mt) and end-of-year seasonality will dominate, placing the outcome in the 50–57 Mt band (bin_9) with high probability. The brief’s high-evidence scenario explicitly assumes normal operations carry that baseline forward.
Counter-evidence found (or absence)
The labor-dispute evidence directly contradicts the normal-operations premise: ETU workers voted 95 %+ YES with 95 % turnout, four unions aligned, unions stated a strike was “very likely” before 30 June, BHP itself described the risk as capable of halting the hub, and no settlement existed as of 1 June (sources 7, 12, 21, 23, 26, 31). This is substantive, multi-source counter-evidence that the brief tags only “medium” strength yet materially under-weights relative to the historical throughput data.
Distribution implications
The strike scenario (medium evidence strength) receives more mass than a naïve consensus would assign, shifting probability from bin_9 into bins 5–8 (40–50 Mt). The low-evidence severe-disruption tail receives only marginal weight, while the high-evidence normal-operations case still anchors the upper bins but is discounted enough that bin_9 receives only 32 % rather than the 55 %+ a pure historical extrapolation would imply.
Strategy: logit_mean • Drafts: 3/3
anchoring_v1The draft’s p50 sits near 53 Mt (the May 2026 level-forecast model anchor), yet the brief provides two strong reasons why the modal outcome should be below that: (1) the level-forecast model was explicitly built by anchoring at May 2026’s 53.1 Mt with zero drift, then treating that as the p50 — this is a classic “use the most recent extreme as the central estimate” anchoring error when May 2026 was itself a record high; and (2) the active labor dispute (95%+ YES strike vote, four unions aligned, “very likely” before June 30 per sources 12, 21, 31) is a named downside risk that the model mechanically excludes. The brief’s own inside-view analysis assigns ~40% probability to meaningful disruption, which should lower the modal expectation from ~53 Mt toward ~48–50 Mt. Additionally, the inside-view placed 35% mass in above 50 Mt (bin_9) while the outside-view variant placed 73% there — this huge variance suggests the reconciler’s logit-mean was dominated by the optimistic outside-view anchor. A flat-prior reading of June 2025 (54.6 Mt), June 2024 (50.4 Mt), and a 40% strike discount centers the modal expectation around 48–51 Mt, not 53 Mt, making bin_9 modestly over-weighted and bins 7–8 (45–50 Mt) under-weighted.
ceiling_v1Top structural concern: the draft places nearly half (49%) of probability mass in >50,000,000 (bin_9) even though the brief itself documents a clear physical/historical ceiling near ~54.6 Mt — June 2025 = 54,584,776 mt (source 22) — and port expansion targets (660 Mt/year) imply an average monthly throughput of 660/12 = 55 Mt (source 30). Bin_9 as used by the drafter therefore conflates modest >50 Mt outcomes (50–56 Mt) with extreme, physically implausible outcomes (e.g., >63–70 Mt) and overweights high tails that the structural facts do not support. Also, the PPA’s month-assignment rule (by vessel departure date) creates strong path-dependence around the June 30 cutoff that the current continuous spread fails to capture — a late-month disruption or scheduling bump should shift mass toward the month-adjacent bin(s), not uniformly widen the top bin.
math_v1The draft’s prediction map sums to 1.000 and its bin allocation is internally coherent, but the distribution shape contradicts the forecaster’s own probabilistic reasoning about strike risk. The rationale cites a ~40% probability of strike-driven disruption (shifting mass into the 35–50 Mt range) but allocates 49% of mass to >50 Mt (bin_9), which under-represents that medium-evidence downside scenario. For consistency with “P(meaningful disruption) ≈ 40%,” bin_9 should receive approximately 30–35%, not 49%. The forecaster explicitly states the strike scenario “shifts probability mass downward” yet fails to implement that shift proportionally.
Reviews consumed: 3
The anchoring reviewer flagged that the draft’s ~49% in bin_9 is dominated by the optimistic outside-view variant (73%) and underweights the explicit ~40% strike risk from a 95%+ YES vote; I partially accept this and shift mass from bin_9 down into bins 5–8. The ceiling reviewer (high confidence) made an important structural point: >50 Mt (bin_9) has limited physical headroom — June 2025’s 54.6 Mt is roughly the realized ceiling and 660 Mt/yr capacity implies ~55 Mt monthly average, so bin_9 should not balloon AND 47.5–50 Mt (bin_8) deserves more weight as the modal packet given May 2026=53.1 Mt and the strike discount. The math reviewer argued in the opposite direction — that the inside-view’s stated 40%/60% disruption split implies bin_9 should be ~32–38% OR (alternatively) up to ~50% if normal-ops mass concentrates there; I reject the upward push to 0.5 because it ignores the ceiling reviewer’s structural argument and the labor dispute. Combining anchoring (shift_lower) + ceiling (redistribute toward bin_8, cap bin_9) yields a distribution with bin_9 around 0.33 and a strong bin_7/bin_8 mode. The contrarian variant (32% bin_9) and anchoring reviewer’s independent forecast (24.5%) bracket this; I land slightly higher at ~33% to respect the high-evidence normal-operations baseline while honoring both downside (strike) and upside-ceiling concerns.
[math] Bin_9 should be raised to ~0.5 to match the 60% normal-ops scenario weight concentrated above 50 Mt.
[ceiling] Path-dependency around the June 30 vessel-departure cutoff warrants overweighting outcomes just below the cutoff.
| Bin | Probability |
|---|---|
| Below 30000000 (bin_0) | 0.6% |
| 30000000-32500000 (bin_1) | 1.0% |
| 32500000-35000000 (bin_2) | 2.2% |
| 35000000-37500000 (bin_3) | 4.0% |
| 37500000-40000000 (bin_4) | 6.5% |
| 40000000-42500000 (bin_5) | 9.0% |
| 42500000-45000000 (bin_6) | 12.0% |
| 45000000-47500000 (bin_7) | 15.0% |
| 47500000-50000000 (bin_8) | 17.0% |
| Above 50000000 (bin_9) | 32.7% |
wooley_rhino • run 878097c1 • $2.83 • 6m29s
Extracted Percentile Values: {0.1: 29200000, 1.0: 37700000, 5.0: 41900000, 10.0: 43600000, 15.0: 44600000, 20.0: 45300000, 25.0: 45900000, 30.0: 46400000, 35.0: 46900000, 40.0: 47350000, 45.0: 47650000, 46.0: 47730000, 47.0: 47810000, 48.0: 47890000, 49.0: 47970000, 50.0: 48050000, 51.0: 48130000, 52.0: 48220000, 53.0: 48310000, 54.0: 48400000, 55.0: 48490000, 60.0: 48850000, 65.0: 49150000, 70.0: 49400000, 75.0: 49650000, 80.0: 49900000, 85.0: 50150000, 90.0: 50550000, 95.0: 51250000, 99.0: 52600000, 99.9: 55100000}
Model’s Answer: My central view is that June 2026 Port Hedland iron ore exports will most likely land a bit above April 2026’s 46.3 million tonnes, but not at an extreme high. The strongest anchors are the recent official PPA-related data points in the research: record 2025 annual iron ore exports around 575 million tonnes, December 2025 at 50.9 million tonnes, and April 2026 at 46.3 million tonnes. Those suggest a high-capacity system that has been running very strongly, with a normal monthly zone in the high-40 millions rather than low-40s.
Step one is anchoring on the recent base rate. A 575 million tonne year implies about 47.9 million tonnes per month on average. June is normally a favorable shipping month relative to cyclone-season months, so absent unusual disruption I would expect something around or modestly above that average, not below April by much.
Step two is adjusting for current conditions. Weak iron ore prices and softer Chinese demand are a mild drag on the upside, but large Pilbara producers are low-cost exporters and usually keep volumes high rather than cutting heavily. Producer guidance in the research remains broadly steady. That argues for only a modest downward adjustment to the “normal dry-season” baseline, not a collapse.
Step three is incorporating the industrial action risk. The Reuters reporting is the most important downside-specific new information. But the timing matters a lot: the threat is for action by the end of June, not clearly from the start of the month. That means the most likely strike scenarios either do not happen in June at all, or only affect the last few days of the month. Because the resolution is for the whole month, that reduces expected impact versus a headline reading of “possible strike.” I still widen the lower tail because even a few days of disruption at BHP could matter, and there is some chance of escalation or indirect effects on maintenance and loading efficiency.
Step four is thinking about tails. Very low numbers would likely require something more than soft demand: a substantial strike beginning earlier than expected, a serious rail or port outage, or an accounting/month-assignment quirk that pushes shipments into adjacent months. Very high numbers would require dry-season smooth operations, strong vessel scheduling, and miners choosing to keep exports elevated despite weak prices. Because 2025 was a record year and December 2025 was already above 50 million tonnes, I want a meaningful right tail above 50 million tonnes, but not an aggressive one.
The assumptions that matter most are these. The first is that the threatened labor action, if it occurs, is late-June or limited rather than an early, severe disruption. I think there is about a 25 percent chance this assumption is materially wrong. If it fails, the distribution shifts down by roughly 2 to 5 million tonnes. The second is that April 2026’s 46.3 million tonnes is a decent recent baseline for the system and not a sign of a more persistent weakening in monthly exports. I assign about a 20 percent chance this is materially wrong. If it fails on the downside, June could be in the mid-40s even without a strike; if it fails on the upside, June could get back into the low-50s. The third is that June sees no unusual non-labor operational shock such as a major rail interruption, berth outage, or rare weather problem. I put about a 10 percent chance on that assumption failing. If it fails, the lower tail gets much fatter.
A low outcome would come from a bigger-than-expected BHP disruption, especially if industrial action starts before month-end or if contingency staffing is ineffective, combined with soft demand and some slippage into July accounting. I think a genuinely low outcome, meaning below about 43 million tonnes, is plausible but still clearly a minority case, around high single digits.
A high outcome would come from clean dry-season operations, no meaningful strike in June, strong vessel line-up management, and miners continuing to maximize throughput despite softer prices. I think an outcome above 50 million tonnes is more plausible than a true crash, but still not the base case, roughly low-to-mid teens.
Overall, my median is a little above 48 million tonnes, with a distribution centered in the upper-40 millions, a wider-than-normal lower tail due to the labor dispute, and a meaningful but not dominant upper tail above 50 million tonnes.
Percentile .1: 29200000 Percentile 1: 37700000 Percentile 5: 41900000 Percentile 10: 43600000 Percentile 15: 44600000 Percentile 20: 45300000 Percentile 25: 45900000 Percentile 30: 46400000 Percentile 35: 46900000 Percentile 40: 47350000 Percentile 45: 47650000 Percentile 46: 47730000 Percentile 47: 47810000 Percentile 48: 47890000 Percentile 49: 47970000 Percentile 50: 48050000 Percentile 51: 48130000 Percentile 52: 48220000 Percentile 53: 48310000 Percentile 54: 48400000 Percentile 55: 48490000 Percentile 60: 48850000 Percentile 65: 49150000 Percentile 70: 49400000 Percentile 75: 49650000 Percentile 80: 49900000 Percentile 85: 50150000 Percentile 90: 50550000 Percentile 95: 51250000 Percentile 99: 52600000 Percentile 99.9: 55100000
Port capacity, long‑term mine scale, and berth and channel constraints impose a persistent mid‑50s million‑tonne June operating envelope.
Seasonal maintenance windows and shipping cycles concentrate throughput into the June quarter unless disrupted by weather or labor events.
Recent monthly loadings have been below the peak but major producers’ guidance and scheduled arrivals support a recovery into June.
Earlier cyclone disruption tightened vessel queues temporarily, but subsequent catch‑up loadings have restored near‑normal berth utilization.
A late‑June electrical or labor stoppage could shift substantial tonnage into July and materially lower the June reported total.
Effective contingency measures, higher vessel arrival rates, or faster cargo completions can re‑establish mid‑50s throughput within days.
The chief uncertainties are timing and scope of industrial action, the composition of the June vessel queue, and month‑end accounting cutoffs.
Small changes in visit completion rates and a few delayed sailings create a narrow but consequential downside tail for the official monthly number.
Forecast (June 2026 Port Hedland iron ore exports): ~51.2 million tonnes (median/P50 not provided)
What will Port Hedland iron ore exports be in June 2026?
Key figures
Historical context
Port Hedland exhibits a powerful 12-month seasonal cycle where June is almost always the peak export month. This is driven by the ‘June flush,’ where Australian miners (BHP, Fortescue, Roy Hill) maximize shipments to book revenue before the June 30 fiscal year-end. Historically, June volumes have climbed steadily: 51.93 million tonnes in June 2023, 54.19 million tonnes in June 2024, and 54.58 million tonnes in June 2025. This pattern typically results in June being roughly 15.5% higher than the average of other months. However, early-year volumes are often disrupted by the cyclone season (November-April), as seen in early 2026 when Cyclones Mitchell and Narelle caused significant shutdowns. Labor disputes at the port are infrequent but high-impact, given the integrated nature of the rail-to-port logistics chain in the Pilbara.
Tailwinds
Headwinds
Detailed reasoning
My analysis for June 2026 Port Hedland iron ore exports centers on the conflict between exceptionally strong historical seasonality and significant near-term operational risks. Historically, June is the highest-volume month for the port as major producers BHP, Fortescue, and Roy Hill push to meet Australian fiscal year-end shipping targets. Data from 2023, 2024, and 2025 shows June volumes of 51.9M, 54.2M, and 54.6M tonnes, respectively. However, several factors suggest June 2026 may struggle to reach these recent heights.
First, the threat of industrial action is high. The Electrical Trades Union (ETU) is balloting approximately 200 workers at BHP’s Port Hedland operations, with strikes ranging from 15 minutes to 24 hours described as ‘very likely’ by the end of June. While BHP maintains contingency plans, even minor disruptions at the world’s largest bulk export hub can create logistical bottlenecks that prevent the port from hitting its theoretical maximum throughput.
Second, the run-rate heading into mid-2026 has been inconsistent. While January 2026 was strong at 49.2M tonnes, subsequent months were impacted by Severe Tropical Cyclones Mitchell and Narelle, which pulled February volumes down to 40.0M tonnes and kept March and April near the 46.3M tonne mark. While there is usually a ‘catch-up’ effect following weather disruptions, the lower baseline in the first four months of the year suggests that achieving a 50M+ tonne month would require a significant and uninterrupted operational surge.
Third, proxy data from PortWatch (AIS satellite tracking) suggests a potential cooling in May 2026, recording a drop in activity. Although the official Pilbara Ports Authority (PPA) figures are the only data used for resolution, these early indicators of maritime traffic cannot be ignored.
I have weighted the forecast toward the upper end of the 40M-50M tonne range to account for the fiscal year-end push, but I have maintained a wide interval to reflect the volatility introduced by potential strikes and the recovery from a heavy cyclone season. The most likely prediction sits just below the 50M mark (49,900,000 tonnes) to reflect a high-performance month that is nonetheless constrained by labor friction and the trailing effects of early-year disruptions. There remains a significant ‘upside’ tail where exports could exceed 50M tonnes if labor peace is maintained and weather remains clear.
Key uncertainties
Conclusion