Part 3 of our fuel shock coverage. Part 1: Middle East Crisis: What the Fuel Shock Means for South Africa’s Cold Chain. Part 2: A River Full of Crocodiles.
The River Is Bursting Its Banks: Seven Weeks On from the Crocodiles
A note before we begin
Between 5 and 7 May 2026, an intense cut-off low-pressure system has moved across South Africa. The South African Weather Service has issued Orange Level 8 disruptive rainfall warnings — the highest tier in active use — for parts of the Eastern and Western Cape. Rainfall accumulations of 100 to 200 millimetres in 24 hours have been forecast for the Garden Route. Four people have died in flooding-related incidents at the time of writing, including a 12-year-old in Taung and a person crushed by a falling tree in an informal settlement in Germiston. KwaZulu-Natal is on Level 6 alert. The N2 and N3 are being monitored for flooding, debris, and possible disruption.
This article is about a different river — the structural diesel and energy position the cold chain industry now depends on. But the literal one demands acknowledgement first. The metaphor we used seven weeks ago has caught up with the reality faster than anyone expected, and the people in Koukamma and Kouga, in the Garden Route, in low-lying KZN, and in the informal settlements that took the worst of it deserve better than to be written through.
With that said.
The river is bursting its banks
On 20 March 2026, this site published A River Full of Crocodiles. The argument was that South Africa was watching one visible threat — the April fuel shock — while five others were closing on the same bank: the Food Crocodile, the Energy Crocodile, the Money Crocodile, the People Crocodile, and the Silence Crocodile.
Seven weeks later, the river is bursting its banks.
This is not the article we expected to write next. The original plan — drafted before the cut-off low system arrived — was to revisit the Energy Crocodile alone and go deeper on the diesel-specific structural position. That article was already overtaken by events when literal storms started taking out roads and lives along the same N2 corridor that carries imported diesel from Durban to the inland market.
So this is a sharper piece than originally intended. It updates one specific question: in the seven weeks since Crocodiles was published, what has actually happened to the diesel position, what new evidence is on the table, and what is the operator-level mechanism by which structural exposure converts into business failure? Three things have changed, one has not, and one was always going to be the test.
What has changed in seven weeks
The Cresco Group research note. A primary-source piece of analysis from a South African energy advisory firm has been published, titled Diesel: South Africa’s last line of defence for Energy Security — now at risk. It is the most thorough public examination of the diesel position to date. It quantifies what Crocodiles gestured at: Eskom’s open-cycle gas turbines absorb 3 to 10% of national refined diesel in stable periods and 20 to 30% under sustained load shedding, and approximately 57% of South Africa’s refined diesel imports transit the Strait of Hormuz. The analysis is the underlying source for several of the recent media pieces on diesel risk that have appeared since.
Eskom’s diesel consumption has collapsed — for now. On 3 April 2026, Eskom reported FY2026 OCGT diesel spend of approximately R6.4 billion — a 62.4% year-on-year reduction from R17 billion in FY2025, and a 76% reduction from R26.6 billion in FY2024. The OCGT load factor for the year was 3.61%. South Africa went 322 consecutive days without an interruption to electricity supply.
This is the good news. It is also the trap. Eskom’s diesel demand is not gone. It is dormant. The OCGT plants at Ankerlig, Gourikwa, and the IPP peakers have not been decommissioned. The capacity to absorb 30% of national refined diesel under sustained load shedding still exists. The Generation Recovery Plan has produced real, documented improvements in plant availability — and the system has used the breathing room to defer the harder structural questions about diesel reserves, refining capacity, and import exposure that Crocodiles named in March.
A Gauteng Long-Haul Operator — current to this week. On 8 May 2026, a Gauteng-based refrigerated carrier wrote to its clients announcing a 7.5% general rate increase, with a 10% surcharge for collections and deliveries to outlying areas. The letter contained two operational data points that are not normally made public.
The first: fuel currently accounts for approximately 35% of operating overheads.
The second: in many cases, the carrier only recovers increased fuel costs through its income cycle some 60 days later.
These figures are central to what follows. They are not unusual for the industry. Dry road freight typically runs at 25 to 28% fuel as a proportion of operating cost. The 7-to-10-point gap is the cost of the Transport Refrigeration Unit — the diesel engine on the trailer that runs independently of the prime mover and powers the cooling system. A refrigerated long-haul truck running Cape Town to Johannesburg with a multi-temperature load consumes around 350 litres of diesel for the prime mover and a further 60 to 80 litres for the TRU on a single round trip. Industry benchmarks place TRU consumption at 0.5 to 1.0 gallons per hour of runtime. The 35% number reflects the structural reality of running two diesel-burning machines in parallel for the duration of every load.
The 60-day recovery lag is the more important number, and it is what this article is really about.
What has not changed
The structural position. Crocodiles documented it in detail and the position has not improved.
Domestic refining capacity is still less than 35% of national demand, down from approximately 80% a decade ago. Sapref, acquired by the Central Energy Fund in May 2024, has not produced a barrel of refined product in either of the two years since the acquisition. There is no public timeline for restart. The Clean Fuels 2 deadline of 1 July 2027 is now closer than the Sapref acquisition date.
Strategic fuel reserves are still estimated at approximately 7.7 to 8 million barrels — about two weeks of national consumption, against a 90-day International Energy Agency benchmark and a 60-day legal requirement. This was true in March. It is true in May. Nothing has been added.
The 57% of South Africa’s refined diesel that ships through the Strait of Hormuz is still routed through the Strait of Hormuz. The Middle East conflict has not closed it, but it has not normalised either. The Dangote 12-month supply arrangement that government has been pursuing is still being pursued. India, the largest single source of South Africa’s refined product imports, is still reallocating barrels to the highest-paying market.
Sasol’s Secunda synfuels operation — approximately 150,000 to 160,000 barrels per day, the world’s only commercial coal-to-liquids facility — is still the swing producer holding the inland market together. Sasol’s Q3 FY26 update, published 24 April 2026, confirms Secunda is currently performing strongly: improved coal quality from the new destoning plant, full-capacity Natref operation, and fuel sales guidance revised upwards from 5–10% to 10–15% above prior year on post-Hormuz demand. This is good news that should be reported clearly. It is also conditional. Sasol is still investing only what is required to keep the existing units running. There is still no replacement programme of equivalent ambition in development. Strong current performance does not change the structural position. It buys time within it.
The system has not moved. Seven weeks ago, the river was full of crocodiles. Today, the same crocodiles are in the same water. What has changed is that the water itself is now rising, both literally and operationally, and the operators standing on the bank are in worse working-capital condition than they were in March.
The 60-day trap
This is the part Crocodiles did not write, because it required an operator-level data point we did not yet have.
Consider what happens to a refrigerated carrier when the diesel price moves up by a rand a litre, sustained.
The carrier pays for the increased fuel today. The carrier’s income cycle is 30 to 60 days, depending on the customer. The 7.5% rate increase the carrier negotiates with its clients takes effect prospectively — meaning it applies to loads collected after the agreed date, not retrospectively to loads already moved. The customers themselves take 30 to 60 days to pay. The carrier therefore absorbs the diesel price increase out of working capital for somewhere between 60 and 120 days before recovery begins to flow through.
In a stable price environment, this is a manageable lag. In a rising price environment, it is not. Each upward step in the diesel price requires a fresh round of customer notification, a fresh negotiation, a fresh implementation date. By the time the rate increase is in effect, the next price step has already happened. The carrier is permanently under-recovered, with the gap funded out of cash on hand.
This is what decapitalisation looks like in cold chain. Operators do not go bankrupt because they cannot recover the price of diesel. They go bankrupt because they cannot recover it fast enough. The 60-day lag between paying for fuel and being paid for the load that consumed it is the mechanism that converts a price spike into a working-capital crisis, and a working-capital crisis is what closes a refrigerated carrier in a low-margin industry.
It is worth being precise about this, because the failure mode is not what most outside observers expect.
Cold chain operators do not run out of diesel. They run out of cash before the diesel runs out. The trucks keep moving and the cold rooms keep cooling — for a while. The genset at the distribution centre keeps starting on cue. The TRU on the trailer holds temperature. From the customer’s side, nothing visible changes. Then a fuel supplier requires payment in advance instead of 30 days. Then a second supplier follows. Then the carrier defers vehicle maintenance to free up cash. Then a TRU compressor fails on a load. Then the load arrives at the DC three degrees out of spec and a retailer rejects it. Then the carrier’s insurance excess on rejected loads rises. Then the bank calls.
None of this is visible to the customer until the carrier closes, at which point the cold chain capacity that company represented disappears from the market with a week or two of notice and a queue of competitors who all have the same exposure to the same diesel price.
The 7.5% rate increase that the Gauteng Operator letter announced is not a price increase the carrier chose to charge. It is the carrier already absorbing several months of cost movement and beginning the long, lagged process of trying to recover it. The next diesel price step — whenever it arrives, from whatever cause — starts the cycle again, while the previous one is still being recovered.
This is the operator pain that is current in May 2026. It is not a forecast. It is the arithmetic of every refrigerated carrier in South Africa today.
The geography proves itself
Crocodiles made the geographic argument in passing. The cut-off low system has made it concretely, this week.
South Africa’s Petronet network — operated by Transnet Pipelines — carries 100% of the country’s bulk inland petroleum and moves more than 16 billion litres of refined product per year across approximately 3,800 kilometres of high-pressure line. Inland Gauteng, where most of South Africa’s perishable freight transits, depends on three sources: the New Multi-Product Pipeline (NMPP) from Durban, the original 555-kilometre Durban-Johannesburg Pipeline (DJP) commissioned in 1965, and Sasol’s Secunda synfuels operation feeding directly into the inland accumulation system.
The arithmetic is tighter than is generally appreciated. The NMPP trunk line currently delivers approximately 148 million litres of product per week — roughly 21 million litres per day — across all grades (petrol, diesel, jet fuel). Designed for ramp to 3,000 cubic metres per hour through additional pump stations, the line currently operates at 1,000 cubic metres per hour. The 60-year-old DJP, described in trade press as “running at full capacity and nearing the end of its design life” as far back as 2013, is still required to operate alongside the NMPP to meet inland demand. Secunda contributes approximately 150,000 to 160,000 barrels per day of synthetic fuel production, of which roughly 35% is diesel — about 8 million litres per day, refined directly into the inland market without ocean-shipping or pipeline exposure to imports.
That is the inland fuel system. Three sources, two pipelines (one of them six decades old), and one synfuels operation that Sasol has confirmed is currently performing strongly — boosting fuel sales guidance from 5–10% to 10–15% above prior year on the back of post-Hormuz demand. The system works in May 2026 because Secunda is performing. The question of what happens when Secunda stumbles is the topic of a forthcoming standalone piece on this site, because it is too large for a paragraph.
When the pipeline-plus-Secunda system cannot meet demand — for any reason, including planned maintenance, contamination events, or capacity constraints — the gap is bridged by road. That road is the N3.
This week, the N3 and N2 corridors are being actively monitored for flooding, debris, and disruption. KZN’s Provincial Disaster Management Centre has activated an emergency Joint Operations Centre. Major transport routes are flagged as exposed. None of this has, at the time of writing, produced sustained closures. The network is holding. The point is not that it has failed — it is that the redundancy on which the inland market depends has been tested by literal weather, in addition to the structural pressures the Crocodiles piece described.
A cold chain operator in Boksburg or Pretoria collecting diesel today is collecting fuel that has either been refined at Secunda, pumped 555 kilometres up the escarpment from Durban via NMPP or DJP, refined at Natref in Sasolburg, or trucked up the same N3 that disaster management is currently monitoring. There is no fifth option. The geography that Crocodiles described as a structural concern is now a current operational variable.
Winter has arrived
The original Cresco scenario analysis modelled the compounding risk of load shedding, Hormuz disruption, and refining tightness arriving together. The framing assumed winter would test the system. Winter has arrived earlier than expected, and the test is happening before the diesel position has been hardened in any structural way.
Three things happen between May and August in southern Africa. Cooking and heating loads on the Eskom grid rise, particularly in the morning and evening peaks. Coal plant performance is more variable as units cycle through maintenance. Renewable generation drops — solar capacity factors fall, and wind, while seasonally complementary, is not yet at the scale to backstop a generation gap.
Eskom’s Summer Outlook in September 2025 forecast no load shedding through to 31 March 2026. That forecast was met. The Winter Outlook was, at the time of writing, due in April 2026. The system has, by every available measure, performed better than it did in 2023. The Generation Recovery Plan has produced real, documented improvements.
The point is not to predict a return to load shedding. The point is that the system has no margin for the combination of events the Cresco analysis describes — and the literal weather currently bursting the literal river is, on its own terms, a partial test of that margin. The cut-off low has not affected the grid in a meaningful way. The next cold front, or the one after, may behave differently. Coal plants under cold-weather stress, wet coal, transmission lines under wind load, hydro generation responding to either too much or too little rainfall — none of these are abnormal in a southern African winter. They are, however, exactly the conditions under which dormant OCGT capacity is dispatched.
If that dispatch happens at any scale this winter, the Cresco scenario stops being a scenario. It becomes a market position that operators are paying for in real time, while still trying to recover the previous month’s fuel cost from customers on a 60-day cycle.
What operators should already be doing
This article is structural analysis, not a checklist, and Crocodiles already laid out an eight-point operator response. The seven-weeks-on update is narrower.
- On the working-capital position: The 60-day trap is real. The contractual response — formal fuel-surcharge clauses tied to the basic fuel price or a published index — is the most consequential change a carrier can make right now. Standard South African road-freight contracts treat fuel as the carrier’s risk. Cold chain contracts mostly inherit that default. They should not. A surcharge clause moves the lag from months to days. Operators with a documented framework activate it on price moves. Operators without one negotiate from weakness while their fuel accounts settle on the new price. The window between today and the next material price step is the window in which this gets done, or doesn’t.
- On the diesel storage question: The R638 regulations on hazardous goods storage are restrictive but not prohibitive. The question for an operator is whether the capital cost of compliant on-site diesel storage is lower than the working-capital cost of repeatedly absorbing a 60-day fuel-price lag. In May 2026, for many operators handling sustained price volatility, the answer has changed.
- On the cold-hold modelling: Operators who modelled their facility’s cold-hold time against genset runtime during the 2023 load shedding peak have the data. Operators who did not are guessing in May, and will continue guessing through winter.
- On insurance: Most refrigerated carriers and cold-storage operators carry insurance against cooling failure. Far fewer carry insurance against fuel-access failure. The product loss in either scenario is identical. The contractual position is not. This is worth a conversation with a broker before winter.
The operational responses to this convergence are not exotic. They are insurance, contracts, storage, and modelling. None is novel. All are standard practice in industries with greater regulatory attention to fuel security — pharmaceutical manufacturing, hospital operations, mining. Cold chain has been treated as a downstream consumer of road freight risk. It is not. It is its own risk category, with its own failure mode, and the operators who recognise this in the next sixty days are better positioned than those who do not.
The river
The Crocodile metaphor was always meant to be uncomfortable. The river was always full. The point in March was that watching one threat is not the same as managing the system.
Seven weeks later, the same crocodiles are in the same water. Eskom’s diesel demand is dormant but uncapped. The strategic reserves are still at two weeks. Sapref is still idle. Hormuz is still a route, not an alternative. Sasol Secunda is older, not newer. The cold chain operators carrying the load are sixty days further into a working-capital position they did not choose and cannot easily exit.
What is new is that the river is now bursting its banks, both metaphorically and literally. The same N2 and N3 corridors that carry inland diesel are being assessed for flood damage this week. The same operators absorbing structural diesel cost are absorbing weather-related route disruptions on top. Winter has not arrived politely, on schedule, in June. It arrived early, with a Level 8 warning, and the grid has not yet been tested by it.
The thesis of the Crocodiles piece was that the convergence — not any individual stress — was the argument. The thesis of this article is narrower. It is that seven weeks of breathing room produced no structural improvement to the diesel position, the operator-level pain has accelerated rather than eased, and the seasonal window for the system to be tested has opened earlier than expected.
The crocodiles are still in the water. The water is rising. The operators who are still standing in twelve months will be the ones who used the time to build pricing mechanisms, fuel storage, and cold-hold resilience while the rest of the industry was watching the visible head and waiting for the river to settle.
That is the question for the next sixty days. Not whether the crocodiles are in the water. They are. Not whether the river will rise. It is rising. The question is whether the cold chain industry is built to outlast the convergence of both.
Sources & References
About these sources
This article draws on primary energy-sector sources (Cresco Group, Eskom, Transnet Pipelines, Sasol’s published Q3 FY26 update), recent South African media coverage of the May 2026 floods, and operator-level data anonymised from a current industry communication. Where projections and analysis extend beyond published data — particularly in the inland fuel system math — the article shows its working and presents figures as analytical positions rather than precise published statistics.
Currency note
Strategic fuel reserve estimates, refining capacity status, Eskom diesel consumption, and Sasol operational position reflect publicly reported positions as of April–May 2026. Weather warnings and N2/N3 disruption status reflect the active cut-off low system of 5–7 May 2026 and may resolve before time-sensitive operational decisions are taken. Readers should verify current status against the relevant operator’s most recent disclosure.
Floods and weather (May 2026)
- Level 8 disruptive rainfall warnings issued for Eastern and Western Cape, South Africa — The Watchers, 5 May 2026 — SAWS warning detail, rainfall forecast 150–200mm.
- Four dead as floods and severe storms batter South Africa, KZN on high alert — East Coast Radio, 6 May 2026 — Confirmed deaths, KZN escalation, infrastructure exposure.
- KZN on high alert after Level 6 weather warning — IOL, 4 May 2026 — Provincial Disaster Management Centre activation, N2 and N3 monitoring.
- Storm systems wreak havoc on Garden Route and parts of Eastern Cape — IOL, 6 May 2026 — Garden Route flooding detail, school closures, infrastructure damage.
Diesel and energy security (March–May 2026)
- Diesel: South Africa’s last line of defence for Energy Security — now at risk — Cresco Group, March 2026 — Primary research note: Eskom OCGT 3–10% / 20–30% diesel share, 57% Hormuz exposure, compounding-shock scenario analysis.
- Eskom delivers sustained grid stability and 62.4% year-on-year reduction in diesel usage — Eskom, 3 April 2026 — FY2026 OCGT load factor 3.61%, R6.4 billion diesel spend, 322 days uninterrupted supply.
- Eskom Data Portal — OCGT usage — Eskom — Public-facing weekly OCGT load factor and diesel expenditure data.
Strategic reserves and refining (current position)
- South Africa holds just 8 million barrels in strategic fuel reserves — IOL, 25 March 2026 — Reserve position confirmed against capacity gap.
- South Africa’s state-owned CEF inks deal for idled SAPREF refinery — Oil & Gas Journal, May 2024 — CEF acquisition of Sapref and SAPIA Q1 2024 import-dependency figure (60% finished fuel imports).
Pipeline and inland geography
- Transnet Pipelines (Our Divisions) — Transnet — Network description, throughput volumes, intake stations.
- Transnet Pipelines (TPL) — Operations — Transnet — NMPP trunk line capacity 148 Ml per week (D10, D50, ULP 93/95, jet fuel); Jameson Park inland accumulation facility 170 Ml.
- About Us — New Multi-Product Pipeline — Transnet — 555 km trunk line, 3 million litres per hour design capacity, network and route description.
- Durban/Joburg fuel pipeline opens after years of delays, spiralling costs — News24, 6 October 2017 — NMPP commissioning, capacity, route.
- Transnet fuel pipelines take shape — SAnews — DJP “running at full capacity and nearing the end of its design life” reference.
Sasol Secunda and Natref operational position
- Sasol — Production and sales metrics for the nine months ended 31 March 2026 — Sasol, April 2026 — Q3 FY26 update; SO production 8% higher YoY; fuel sales guidance revised upwards to 10–15% above FY25; post-Hormuz energy security positioning.
- Operational recovery at Secunda anchors Sasol’s half-year result — Moneyweb, 22 January 2026 — Destoning plant (R1bn) operational from December 2025; Natref full-capacity operation despite Prax SA business rescue; FY26 fuel sales guidance lift.
- Sasol launches Destoning Plant to improve coal quality and strengthen operational performance — Sasol, 13 March 2026 — Destoning plant beneficial operation; Q1 FY26 average sinks reduced below 14%.
Cold chain fuel intensity
- How Should Reefer Fleets Tackle the Diesel Double Whammy? — Transport Topics, March 2026 — TRU diesel consumption benchmarks (0.8 gph single-temp, 0.9 gph multi-temp).
- Fuel economy — cutting the cost of running refrigeration equipment — Thermo King — Trailer TRU lifetime fuel consumption.
Operator data point (anonymised)
Private client communication, Gauteng-based refrigerated carrier, dated 8 May 2026 — General rate increase 7.5%, outlying-area surcharge 10%, fuel approximately 35% of operating overheads, ~60-day income-cycle recovery lag.
Operational disclosure
The Curator’s editorial work is independent of the operational businesses that fund this site. The publisher of ColdChainSA also operates The Frozen Food Courier, a refrigerated courier service active in Gauteng and the Western Cape. The operator-level data referenced in this article was sourced from a separate Gauteng-based refrigerated carrier’s communication to its clients in May 2026, with figures preserved and identifying details anonymised. The Frozen Food Courier faces the same structural diesel exposure described in this article, which is one of the reasons it is being written.