Bitumen Supply to Ethiopia: The Djibouti Corridor, Altitude and Grades

Ethiopia became landlocked in 1993, when Eritrea became independent and the ports at Assab and Massawa passed out of Ethiopian territory. It is the most populous landlocked country in the world, and it has run a large and long-standing national road programme. Every tonne of imported binder therefore has the same shape of journey: an ocean leg to a foreign port, a customs frontier, and then a land leg measured in the high hundreds of kilometres before the material reaches an asphalt plant. There is no version of this transaction in which the sea leg is the whole story, and there is no version in which a price quoted to a port is a delivered price.

The corridor structure is unusually concentrated. Djibouti carries the overwhelming majority of Ethiopian foreign trade — a share commonly cited at around ninety per cent or more, a figure given here as the commonly cited order of magnitude and not as a verified statistic — through a road corridor with two alternative frontier crossings and a standard gauge railway running beside it. The realistic alternatives are the Berbera corridor through Somaliland into eastern Ethiopia, and the Port Sudan corridor into the north-west by way of Gondar. Both exist, both are long, and neither displaces Djibouti at the scale of national trade. A buyer choosing between them is not choosing between a short route and a long one; they are choosing between three long ones, and the deciding variables are which part of Ethiopia the project sits in and what the forwarder can actually execute.

The part almost nobody writes: Ethiopia is cold where the roads are

The second fact is the one that ought to change how an offer is written, and it is the one that generic supplier pages never touch. Most of the Ethiopian population, most of its cities and most of its paved road network sit on a high plateau. The traditional Ethiopian classification of land by altitude is not folklore; it is used in the country’s own agricultural and planning literature and it maps almost exactly onto the binder problem:

• Bereha, desert lowland below roughly 500 m.
• Kolla, lowland from roughly 500 to 1,500 m.
• Weyna Dega, midland from roughly 1,500 to 2,300 m.
• Dega, highland from roughly 2,300 to 3,200 m.
• Wurch, afro-alpine above roughly 3,200 m.

Addis Ababa, at around 2,355 m, is a Dega city. So are Dessie at around 2,470 m and Debre Birhan at around 2,840 m. Gondar at around 2,130 m, Mekelle at around 2,080 m, Bahir Dar at around 1,800 m, Adama at around 1,700 m, Hawassa at around 1,700 m and Jimma at around 1,780 m are all Weyna Dega. The consequence is that the temperature a pavement experiences across the settled core of Ethiopia is nothing like the temperature a pavement experiences at Djibouti, at Mombasa, in the Gulf or in the Ethiopian lowlands themselves. The failure mode the plateau is designing against is thermal and fatigue cracking under a large daily cycle, not rutting under sustained heat. A hard grade specified because someone assumed an African market is a hot market is the wrong material, and it is wrong in the direction that produces cracking rather than the direction that produces a rut.

And yet the hot-climate reflex is not simply wrong

What makes Ethiopia genuinely interesting rather than merely counter-intuitive is that the country also contains the other extreme, and contains it close by. The Danakil Depression in the Afar lowlands lies more than 100 m below sea level, with figures of around 125 m below sea level commonly cited for the lowest ground, and the multi-year record taken at Dallol in the 1960s produced an annual mean temperature commonly cited at around 34 to 35 °C, frequently described as the highest annual mean recorded for an inhabited place anywhere on earth. The lower Awash valley, the Somali region lowlands around Gode and Kebri Dehar, and Gambella in the west are the same argument in a milder form. Dire Dawa, a major Ethiopian city on the eastern corridor, sits at around 1,270 m and is hot by any standard.

So Ethiopia is not a cool market and it is not a hot one. It is a market with roughly two and a half kilometres of vertical relief between its lowest inhabited ground and its plateau cities, and a road network that spans the whole of it. That is the honest reason this page will not give a national grade, and it is the reason a supplier who offers one has told you they did not ask where the site is.

The demand pattern underneath

Ethiopian road demand has been shaped for a long time by a national programme approach rather than by a smooth market. The federal network is administered by the roads authority of the federal government — long known as the Ethiopian Roads Authority and reorganised in recent years, with the successor commonly referred to as the Ethiopian Roads Administration — and the long-running Road Sector Development Programme has run in successive phases, with a rural access component that put very large quantities of low-volume road on the ground. Alongside the federal network, Ethiopia’s federal constitutional structure gives the regional states their own roads bureaus, their own budgets and their own procurement, and donor-financed trunk projects are designed and supervised by international consultants working to their own templates.

Two consequences follow for a supplier. First, specification practice is not uniform. A federal trunk contract, a regional roads bureau package, a municipal resurfacing job in Addis and a donor-supervised highway can each carry a different binder clause, a different set of cited test methods and a different view on modified binder. There is no single Ethiopian requirement to quote against. Second, a very large share of the network is surface dressed or otherwise sealed rather than surfaced in asphalt concrete, which pulls cutback and emulsion into Ethiopian enquiries far more often than a pure asphalt-concrete market would.

The five things that decide an Ethiopian order

• The delivery town and its altitude, not the country. Addis at around 2,355 m, Semera in Afar at around 400 m and Gambella in the west at somewhere between roughly 450 and 530 m are three different engineering problems with one national flag over them.
• Which corridor, and therefore which frontier, which transit procedure and which set of handovers.
• Where the goods are cleared — at the frontier, or inland at a dry port, which is the normal Ethiopian arrangement for corridor container traffic and which changes the whole cost structure.
• How the payment is actually going to work. On this market the foreign exchange and banking mechanism is more often the binding constraint than the freight, and it is settled before contract or not at all.
• The packing, chosen from the length of the land leg and from what genuinely exists at the receiving yard, which on an Ethiopian rural project is usually less than a freight quotation implies.

What this page does not tell you

It does not state that any port, berth, railway, road, border post, dry port or corridor service is open today, running to a schedule, equipped for this commodity or available for your cargo. Geography is described because geography is stable and publicly checkable; operating status is not, it changes, and it belongs to the freight forwarder who is accountable for it. Road distances are given as commonly cited approximations for orientation and for nothing else. No transit times, freight rates, vehicle payloads, vessel or tanker capacities, container free time periods, detention charges, duty rates, taxes, levies or exchange rates are stated anywhere. No haulier, forwarder, terminal operator, clearing agent, inspection body, refinery, shipping line or client is named as a counterparty, and no presence, office, agency or shipping history in Ethiopia, Djibouti or any corridor country is claimed. It states no position on whether any product is currently within the scope of any conformity assessment arrangement, and it names no currently appointed inspection body, for the reason set out at length in the conformity section. It quotes no Ethiopian national standard designation for paving bitumen, because the requirement that binds a shipment is the clause the tender incorporates. And it is not legal, customs, regulatory, banking or compliance advice; a buyer trading across these frontiers must take independent advice covering the goods, the parties, each customs territory on the route and the payment mechanism.

The port at the ocean end

Djibouti’s port complex comprises the historic port at Djibouti city and newer facilities at Doraleh, including a container terminal and a multipurpose port, together with specialised terminals for petroleum products, gas and bulk commodities, and free zone facilities alongside them. It is the ocean end of the corridor for the overwhelming majority of Ethiopian trade, and its whole commercial logic is built around transit rather than around Djiboutian consumption.

The inference an Ethiopian enquiry most often makes wrongly is from the oil terminal. Because Djibouti handles petroleum products at scale for onward movement to Ethiopia, buyers assume a bitumen parcel can be discharged in bulk, heated and drawn off. Heated tankage, drum handling equipment, covered storage and the ability to receive a particular packing are contracted commercial arrangements with individual terminal operators. They are never attributes that can be inferred from a port name, from the presence of an oil terminal, or from the fact that a port is large and busy. A terminal engineered for fuels that flow at ambient temperature tells you nothing at all about whether a bitumen parcel can be discharged, heated, stored and released there. That is the practical reason drummed cargo through the container terminals, rather than bulk through a liquid berth, is the format most Ethiopian corridor buyers end up planning around. Confirm what the receiving facility can actually do for your product and your packing, in writing, before a parcel is fixed. Nothing on this page states that any Djiboutian or Ethiopian facility handles bitumen, and no facility named here is a counterparty of ours.

The railway, and the one thing it does not fix

The Addis Ababa to Djibouti standard gauge railway is a genuine structural advantage that the East African corridor through Kenya does not have. It is built to 1,435 mm standard gauge and electrified, is commonly given as roughly 750 km in length, and it runs from the Djibouti area through the frontier at Dewele and through Dire Dawa and the Awash and Adama area to a terminal at Indode near Addis Ababa. It is operated by a joint Ethiopian and Djiboutian railway company. Crucially, there is no break of gauge anywhere on it. The old metre gauge line that historically connected the same two cities is a separate and now largely defunct system, and the two should never be conflated in a routing discussion.

What the railway does not remove is the mode change at each end. A container has to reach the rail terminal at the ocean end and leave it at the Ethiopian end for a road movement to the dry port, the depot or the plant. Three consequences follow for drums. Every lift is a handling event, and handling events are where seams get stressed, closures get knocked and counts get argued about; a packing specification adequate for a single door-to-door road movement is not automatically adequate for a movement lifted several times. Where the box is emptied changes who is holding it and for how long. And a rail movement runs to its own operating pattern rather than to a truck’s, which is an advantage for planning and a constraint for urgency. Whether the railway will carry this commodity in your packing on your date is a forwarder question and this page states nothing about it.

The dry ports, and why Ethiopian cargo is usually cleared inland

This is the piece of Ethiopian machinery most first-time exporters have never met, and it changes the shape of the transaction. Ethiopia operates a network of inland dry ports at which imported cargo is declared and cleared, rather than clearing everything at the frontier. The principal one for Djibouti corridor container traffic is Modjo, roughly 70 km south-east of Addis Ababa on the Adama road, with others including Semera in Afar, Dire Dawa, Combolcha, Woreta, Gelan and Kality in the Addis area.

The practical consequences are worth spelling out:

• The frontier is not the clearance point. The goods typically move from the port to the dry port under a transit arrangement and are entered there. That means the customs event, the inspection event and any conformity check happen several hundred kilometres inland from the sea and a long way from the seller.
• The container goes inland and has to come back. On a round trip of the order of eighteen hundred kilometres with a frontier in the middle, that is not measured in days. Shipping lines allow a contractually agreed period of free time and charge detention after it; the free period and the charge are commercial terms between the buyer and the line and no figures are stated here. Many Ethiopian corridor buyers therefore strip the container at the dry port and move drums onward on flatbed vehicles. That decision has to be made before booking, because it changes the packing plan, the handling count and the insurance arrangement.
• Dwell at the dry port is a storage question, not just a cost question. Drums standing in an open compound in the Awash or Modjo sun for a period nobody planned for are drums with a corrosion and marking problem. Covered storage and short dwell are worth asking about explicitly.

Multimodal and unimodal, which is an Ethiopian decision and not a freight one

Ethiopia has for many years operated a distinctive arrangement for corridor imports in which the movement from the load port all the way to an Ethiopian dry port can be handled as a single multimodal transaction under one transport document, historically the exclusive role of the state logistics enterprise, with private participation opened up progressively in more recent years. The alternative, generally called unimodal, leaves the buyer or their forwarder to arrange the Djibouti to Ethiopia leg separately from the ocean leg.

This page states no position on what is currently permitted, for whom, or on which routings, because the policy has changed more than once. What a seller needs to understand is the shape of the difference, because it decides how the offer should be written:

• Under a multimodal arrangement the transport document covers the whole movement to an inland point, which changes what the bank is being asked to hold and where the seller’s obligation ends.
• Under a unimodal arrangement the ocean leg produces a bill of lading to Djibouti and the land leg produces a separate road consignment note or rail waybill, and the join between the two is a commercial gap somebody has to own.
• The choice affects who selects the inland carrier, who bears standing time and who is exposed to detention on the box.

Ask the buyer which arrangement they are using before quoting, and ask a licensed customs broker in Ethiopia what currently applies. An offer that assumes the wrong one is an offer that cannot be executed as written.

The road itself: a climb, not a haul

The Djibouti corridor is not a flat run. It starts at sea level and finishes on a plateau at around 2,355 m, and it does most of the climbing in the second half after crossing the Afar lowlands. Two things follow. For the carrier, sustained ascending grades with a heavy, dense cargo are a fuel, tyre and scheduling proposition rather than a distance proposition, and long descending grades on the return are a braking one. For the pavement engineer, the places where heavily loaded vehicles move slowly or stand still are exactly the places rutting appears first: climbing lanes on the ascents, weighbridge approaches, the frontier apron, and the truck concentrations around Modjo, Adama and the Addis approaches. That is a modification argument for specific locations, not a national grade argument.

Axle loads and gross vehicle weights on the corridor are controlled under the applicable vehicle load regime and enforced at weighbridges. No payload figure appears anywhere on this page. What a specific vehicle and trailer combination may lawfully carry on a specific alignment, in Djibouti and then in Ethiopia, is a matter for the carrier through the forwarder, and it is two regimes rather than one.

The delivery term has to match the mode, and it has to name a place

Under Incoterms 2020, FAS, FOB, CFR and CIF are rules for sea and inland waterway transport. They are built around a vessel and a port. They are coherent for a parcel discharging at Djibouti and they have no coherent meaning at all for a truck arriving at Modjo or a plant in Bahir Dar. The rules that work for any mode are FCA, CPT, CIP, DAP, DPU and DDP, and every one of them names a place.

Three errors recur on Ethiopian business:

• Quoting CIF Djibouti against an inland Ethiopian project and letting it be read as a delivered price. For a project in Addis, Bahir Dar or Mekelle it prices the ocean leg and leaves the frontier, the transit, the dry port and several hundred kilometres of climb entirely open. Label it as what it is, so the buyer compares like with like.
• Naming a country instead of a place. DAP Ethiopia is not a delivery term. DAP followed by a named dry port, town, plant or project site is. On a corridor movement the named place must also settle who arranges the formalities at the frontier and who bears the cost of a vehicle standing and waiting, because on a land leg the truck and driver are idle at somebody’s expense.
• Agreeing DDP into Ethiopia casually. DDP puts import clearance and charges on the seller. In a market where the importer of record needs a local customs identity, where clearance is normally effected inland at a dry port, and where the settlement of import charges is bound up with the foreign exchange machinery described further down this page, that is a far heavier undertaking than it looks on a term sheet.

Packing, and why drums dominate this corridor

Site standard loading figures, which apply to the sea leg and to any container movement inland, are as follows. New steel drums of 150 kg net give 80 drums and 12 MT per 20 ft FCL. Drums of 180 kg net give 80 drums and 14.4 MT. Drums of 185 kg net give 80 drums and 14.8 MT. Jumbo or poly bags of 1 MT give 20 bags and 20 MT. The 20 ft box is the unit because a cargo of this density reaches its weight limit long before it fills a larger container. Use these figures to fix the drum count, the packing cost per tonne and the number of packages on the transport document; then convert to vehicles with the forwarder rather than with a calculator.

On this corridor the choice between drums, bags and heated bulk is decided by the land leg, and the honest answer is that drums dominate, because the haul is long, the receiving end is usually not equipped, and the cargo is cleared inland rather than at the frontier. The reasons are cumulative:

• Bulk requires heated storage at the destination. A heated road tanker is useful only if the receiver has tankage of adequate capacity, a compatible discharge connection, a pump and the ability to take the whole load promptly. An Ethiopian contractor running a mobile plant on a regional road project several hundred kilometres from Addis has none of that, and no freight quotation changes it.
• A tanker holds temperature badly over nine hundred kilometres with a frontier on it. Time standing at a crossing or at a weighbridge is time cooling, and reheating a stiffened load is an operational problem with a cost and a risk attached rather than a delay.
• Bulk sits awkwardly with an inland clearance model. Where goods move from the port to a dry port under customs supervision and are entered there, the whole procedure rests on the goods that arrive being demonstrably the goods that left. That is not a movement in which product can be conveniently split between two receivers or partly discharged en route.
• A drum fails locally. A damaged drum costs one drum out of eighty. A compromised bulk load costs the consignment, at the far end of a corridor, with no realistic reverse gear.
• Drums allow staged call-off and altitude-tolerant storage. On a plateau project fed at the end of a long corridor, the ability to take material in the quantity the site can use, store the rest under cover through a wet spell and heat it one unit at a time is worth more than the packaging saving.

Where bulk does deserve a hearing is at large fixed installations with permanent heated tankage, and on short local movements from such an installation. That is the honest boundary. Two physical points come with drums on this route. Specify new steel drums and say so in the contract: reconditioned drums are the commonest source of contamination disputes in this trade anywhere, and a cargo passing through a port, a mode change, a frontier and a dry port gives that argument several places to start. And plan the storage environment deliberately, because an Ethiopian consignment can sit in a humid, salt-laden coastal yard at one end of its journey and a fierce Afar or Modjo compound in the middle, both of which work on drum seams and closures faster than most buyers expect.

One regulatory line has to be settled before a tanker is booked rather than after. Where bitumen is offered for carriage above 100 °C it falls to be classified as UN 3257, elevated temperature liquid, n.o.s., Class 9 under the UN model regulations, while packed bitumen moving at ambient temperature is treated differently. Whether that applies to your movement, and what marking, documentation, equipment and driver qualification follow from it in each country on the route, is a question for the carrier and the forwarder and not one this page answers.

The physics, stated plainly

Air temperature falls with height. In the standard atmosphere the environmental lapse rate is taken as about 6.5 °C per 1,000 m, and while real conditions vary, the order of magnitude is reliable enough to reason with. A site at 2,355 m therefore sits roughly fifteen degrees below a sea-level site under otherwise comparable conditions. That single number explains the whole Ethiopian anomaly. Addis Ababa lies at around nine degrees north, deep inside the tropics, and has a climate that a visitor from a temperate country would describe as a mild spring that never ends. Nothing about its latitude prepares a buyer for that, and everything about its altitude does.

Two secondary effects sharpen it. Thin, dry highland air with clear skies radiates heat efficiently at night, so plateau nights cool further and faster than lowland nights at the same latitude, which is why frost is recorded on Ethiopian ground at latitudes where it is unthinkable at sea level. And the same thin, clear air passes more solar radiation by day, so a plateau pavement under direct sun still warms substantially above the air temperature. The result is not a cold pavement. It is a pavement with a large daily cycle: a warm, strongly irradiated surface in the middle of the day and a cold surface before dawn, repeated every day of the dry season.

What that does to the failure mode

Grade selection is really a question about which failure the pavement is being defended against, and altitude changes the answer:

• In a Gulf, Danakil or Awash pavement the binder spends long periods hot and heavily loaded. Permanent deformation dominates. The defence is a stiffer binder at service temperature, which means a harder penetration grade or a modified binder, and the low-temperature end of the specification is not the constraint.
• On an Ethiopian plateau pavement the binder rarely sees sustained extreme service temperature, but it is cycled through a wide range every day, and on clear dry-season nights at the top of the Dega and throughout the Wurch it is taken down to and past freezing. Thermal and fatigue cracking dominate. The defence is a binder with enough relaxation capacity at low temperature, which means a softer penetration grade, not a harder one.

Put the two together and the reason a national grade fails becomes obvious. The same specification cannot be right in Semera and in Debre Birhan, and a road that runs between them changes the answer along its own length. This is not a subtle engineering refinement. Specifying a hard grade for a plateau project because the country is in Africa is a decision that shows up as cracking, and specifying a soft grade for an Afar project because the capital is cool is a decision that shows up as a rut.

The one place the penetration test is unusually honest

There is a well-known awkwardness in penetration grading: the test is run at 25 °C, a temperature that a hot-climate summer pavement does not experience for months at a time, so the grade name describes the binder in a condition almost unrelated to the condition in which it fails. On a Basra or Danakil pavement that criticism is entirely fair.

On the Ethiopian plateau it is not. Twenty-five degrees is close to a normal Addis Ababa daytime maximum. For a Dega or upper Weyna Dega site the penetration test is being run at something like a real service temperature, which makes the grade name more informative here than in almost any other market on this site. That is a small point with a practical edge: on a plateau job, a measured penetration value is worth reading carefully, because it is telling you about the material in a condition the pavement will actually meet.

If the tender is written as a performance grade

A performance grade cannot be derived from a penetration certificate. Where a donor-financed or internationally supervised Ethiopian project specifies to AASHTO M320, or to an MSCR-based grade under AASHTO M332 on heavily loaded pavement, the evidence required is a different family of tests: dynamic shear rheometer testing on unaged and rolling thin-film oven aged binder to AASHTO T315, pressure ageing vessel conditioning to AASHTO R28, bending beam rheometer testing of the aged residue for the low-temperature grade to AASHTO T313, rotational viscosity to AASHTO T316, and where the clause is written against MSCR, multiple stress creep recovery to AASHTO T350. The ASTM counterpart specification for performance-graded binder is ASTM D6373.

Two Ethiopian points about performance grading are worth knowing before you engage with such a clause. First, the performance grade system derives its two designations from pavement temperature: the high-temperature grade from a seven-day maximum pavement temperature and the low-temperature grade from a minimum pavement temperature, both computed from site air temperature statistics. Altitude therefore enters the calculation automatically, through the air temperature record for the actual site. There is no separate altitude correction to apply by hand, and there is no substitute for using the right site’s data: running the calculation on national or regional averages in a country with this much relief produces a grade that is wrong for most of the places it is applied.

Second, and practically: on an Ethiopian plateau project it is the low-temperature half of the designation that is likely to bind, and that is precisely the half an exporter working in the penetration idiom is least likely to hold test data for. Establish at enquiry stage whether the data exists rather than at certification stage. On an Afar or Somali region project the high-temperature half binds instead, and grades at that level commonly require polymer modification rather than a harder straight-run binder.

The construction season, and why it is not the same question as the grade

The dominant highland pattern is a main rainy season, kiremt, running roughly June to September, and a shorter, less reliable rainy period, belg, roughly February to May. The south and south-east run a different bimodal pattern, with rains roughly March to May and again roughly October to November. Bituminous mixes cannot be laid on a wet surface at all, so on the plateau the practical working window runs broadly from October to May, and the main rains close it.

Two consequences that matter more than they look. The first is that the plateau is coolest and cloudiest during the rains, which means the season is bounded at one end by water and at the other by the compaction window: a thin lift at 2,400 m on a cool, breezy plateau evening cools faster than the same lift at sea level, and mat cooling is a genuine construction control here rather than a textbook caution. The second is that a corridor of eight or nine hundred kilometres with a frontier on it cannot be used to correct a late ordering decision. Material for a plateau season has to be moving before the season opens, and drummed material held under cover on site is the normal precaution rather than a contingency.

One more Ethiopian variable: the aggregate

Ethiopian highland geology is dominated by Tertiary volcanic rocks — the trap series basalts, ignimbrites, trachytes and associated scoria of the plateau and the Rift. Volcanic aggregate of this family is entirely usable and is what most Ethiopian projects will use, but some of it is vesicular and absorptive, which affects the binder content a mix actually needs, and adhesion between binder and aggregate varies with the particular source.

The discipline that follows is unglamorous. Affinity between binder and aggregate cannot be certified from a binder sample. No supplier’s certificate, however complete, evidences anything about how the binder will behave against a specific quarry’s stone. Where moisture damage is a concern — and in the wetter west and south-west of Ethiopia it should be — commission the test at mix level with the actual project aggregate: affinity to EN 12697-11, or moisture-induced damage on the compacted mix to AASHTO T283. Any supplier claiming moisture performance on the strength of a binder certificate is overstating what the document can carry.

The idiom, and where it comes from

Where an Ethiopian road project carries a written technical specification, the bituminous section is normally built on penetration grading: the binder is named by a penetration band measured with the needle test at 25 °C, and the mix design and construction clauses sit on the federal roads authority’s own series of design manuals and standard technical specifications. Those manuals are not written in a vacuum. The design reasoning behind Ethiopian and wider East African flexible pavement practice is traceable to the tropical and sub-tropical guidance published by the British transport research laboratory: Overseas Road Note 31, on the structural design of bitumen-surfaced roads in tropical and sub-tropical countries, and Overseas Road Note 3, on surface dressing in tropical and sub-tropical countries. Those two documents explain more about why an Ethiopian specification looks the way it does than any national standard designation would, and the second of them explains why cutback and emulsion turn up so often in Ethiopian enquiries: a very large part of the network is sealed rather than surfaced in asphalt concrete.

Why this page quotes no Ethiopian standard designation

Ethiopia has a national standards body, and it has been through more than one reorganisation. The former Quality and Standards Authority of Ethiopia was restructured, with standards development passing to the Ethiopian Standards Agency, conformity assessment services to the Ethiopian Conformity Assessment Enterprise and legal metrology to the National Metrology Institute of Ethiopia; the institutional arrangements and the name of the standards body have been revised again since. Ethiopia is a member of the African Organisation for Standardisation, and standards published under the national scheme carry Ethiopian Standard designations.

No such designation for paving bitumen is printed on this page, and that is deliberate. The reasons are the same ones that apply on every market page here, with one Ethiopian aggravation. Procurement runs through the federal roads administration, the regional states’ own roads bureaus, municipalities and donor-financed projects designed by international consultants, and their documents are not uniform. The standards institution itself has been renamed and restructured, so a designation quoted from memory may also be attached to a body that no longer exists under that name. And a number quoted into a compliance box on an offer form is a false compliance claim sitting inside a contract. If an enquiry form asks which Ethiopian standard the cargo complies with, the honest and correct answer is that the binding requirement is the one the tender document incorporates, and that you will quote against that clause once you have seen it. Ask for the clause. It is a normal request and a serious buyer will send it.

The first designation trap: 80/100

The grade heard most often in Ethiopian and wider East African conversation, after 60/70, is 80/100. It is worth being precise about it, because it belongs to neither of the two standards an export certificate is usually written against.

• ASTM D946, the standard specification for penetration-graded asphalt binder for use in pavement construction and the reference behind most Middle East export documentation, names the grade 85-100. It contains no grade called 80/100.
• EN 12591, the European standard for paving grade bitumens, names the band 70/100. It contains no band called 80/100.
• The older British standard for bitumens for roads, from which much East African practice descends, named grades in a different style again and has since been withdrawn and superseded by the European standard. A specification assembled from an older template may therefore cite a document that no longer exists in the form the clause assumes.

A clause that says 80/100 is telling you the approximate consistency the engineer wants and is not telling you which requirement table the material will be judged against — and the requirement table is where the flash point limit, the solubility limit, the ductility requirement and the ageing criterion actually live. The arithmetic makes the point sharply. A batch measuring 82 dmm satisfies a literal reading of 80/100 and satisfies EN 12591 band 70/100, but it fails ASTM D946 grade 85-100 outright. A batch measuring 74 dmm satisfies EN 12591 70/100 and fails both a literal 80/100 and D946 85-100. The bands are not synonyms, and treating them as such is how a cargo gets rejected at the far end of a corridor.

The instruction is simple and it belongs in the first reply to the enquiry. Ask which standard’s table the clause intends, and quote in the same words the tender uses. If the tender says 80/100, do not silently substitute 85-100 or 70/100; state on the offer which specification the material is certified to, print the measured penetration, and obtain the engineer’s written approval for any cross-reference before dispatch. A cross-reference table is a basis for a conversation with the engineer. It is never a defence at delivery.

The same care applies one band harder. Bitumen 60/70 is an ASTM D946 grade with a 10 dmm penetration window. The nearest European band is EN 12591 50/70, which is twice as wide and reaches lower. On penetration every 60/70 batch sits inside 50/70, and the converse is false, because a binder correctly supplied as 50/70 may measure anywhere down to 50 dmm and would fail a 60/70 requirement. But penetration is not the only line EN 12591 controls, and this is where the substitution actually breaks: the 50/70 band also carries a softening point range, commonly given as 46 to 54 °C, which sits under the 49 to 56 °C a typical export 60/70 sheet prints. A batch at the upper half of the export band can therefore pass 50/70 on the needle and fail it on ring and ball. Read the direction of that argument carefully before offering one band against the other. One notational point also saves correspondence: AASHTO M20 writes penetration grades with a hyphen — 40-50, 60-70, 85-100 — while the export trade writes 40/50 and 60/70 with a slash. Those are the same bands measured by the same needle test at 25 °C, and a clause naming grade 60-70 is not naming something different from what an exporter calls 60/70.

The second trap, and it is the Ethiopian one: the band is not a point, and here the risk runs the other way

Every market page in this trade warns that a grade name is a band rather than a point, and that two cargoes can both be genuine 60/70 and behave differently. Two measurements make the point: one batch at 61 dmm with a softening point near 56 °C, another at 69 dmm with a softening point near 49 °C. Both are in grade. Both pass a conformity check against the band.

What is specific to Ethiopia is the direction of the risk. In a hot market the buyer’s exposure is a batch at the soft end of the band, which is why those pages tell you to police the top of 60/70 or step to 40/50. On the Ethiopian plateau the exposure inverts. For a Dega site the batch you do not want is the one at the hard end of the band, because it carries less relaxation capacity into exactly the daily cycling the site imposes. A plateau buyer contracting 80/100 who receives a batch at 81 dmm has received material that is in grade and less suited to the job than a batch at 96 dmm would have been. That is the opposite of the advice a generic hot-climate page gives, and it is the single most useful sentence a supplier can put in an Ethiopian offer.

The practical responses, in order of usefulness:

• Require the measured value, not the band, on the Certificate of Analysis. A certificate that reprints the specification range every time tells you nothing about the batch you are buying.
• Agree a narrower contractual window in writing, and set it in the direction the site needs. For a plateau job that means the softer half of the band; for an Afar or Somali region job it means the harder half. The grade name does not change and the tender is still satisfied; what changes is that you hold a contractual right to the material you actually need.
• Read softening point as hard as you read penetration. Ring and ball to ASTM D36 is the line that speaks most directly to behaviour at service temperature, and it is the line most often skimmed on an export offer.
• Add the low-temperature line for a Dega or Wurch site, because no standard export sheet carries one and the plateau is where it matters.

Five questions to ask about any Ethiopian binder clause

• Which document, and which edition? Specifications assembled from older templates carry grades, limits and cited standards that differ from current published texts, and some cite standards or institutions that have since been withdrawn or renamed. Take the acceptance limits from the text the tender incorporates, not from a refinery data sheet and not from memory.
• Penetration, viscosity or performance grade? Penetration is the ordinary Ethiopian case. A clause drafted from an older American viscosity-graded template may name AC-type designations under AASHTO M226; one drafted from Superpave practice will name a performance grade under AASHTO M320 or an MSCR-based grade under AASHTO M332. These are three different measuring systems and a grade in one cannot be inferred from a grade in another.
• Which ageing procedure is controlled? The thin-film oven test (ASTM D1754) and the rolling thin-film oven test (ASTM D2872, AASHTO T240, and in the European system EN 12607-1) are different exposures and their results are not interchangeable. Middle East export certificates carry TFOT by default. If the clause controls RTFOT, add that test to the schedule before the batch is certified, not after the cargo has crossed a frontier.
• Is there a low-temperature requirement, and does the site need one even if the clause omits it? This is the Ethiopian question. An Afar or Gambella project does not need one. A Dega project at 2,300 m and above, with cold clear nights and a large daily cycle, does. Fraass breaking point to EN 12593, or a bending beam rheometer requirement under AASHTO M320, has to be agreed in writing because no export sheet carries it as standard.
• Are prime and tack coat products in the same package, and is a modified binder required? Cutbacks under ASTM D2027 and emulsions under ASTM D977 or ASTM D2397 are separate products with their own acceptance tables, and a paving-grade certificate evidences nothing at all for them. Where polymer modified binder is specified, the European product specification is EN 14023, with elastic recovery to EN 13398 and storage stability to EN 13399.

Adulteration, solubility and the discipline of the handover

An Ethiopian consignment changes hands more often than most: loading point, vessel, port, a mode change to rail or road, a frontier, a dry port, and a final haul on somebody else’s trucks. Each of those is a place where quantity, condition or composition can be argued about, and the argument usually starts several hundred kilometres past the last place either party was standing. The technical defence is unglamorous and effective. Solubility in trichloroethylene to ASTM D2042 is the line that shows whether the material is bitumen or bitumen extended with mineral matter; if a certificate omits solubility, treat that as a finding rather than an oversight. Appoint an internationally recognised inspection company to attend loading, sample across the consignment to ASTM D140, and seal retained samples held by both parties. Then carry that discipline inland: record drum count, drum condition and seal numbers at every handover, because a recorded count is what closes a dispute before it opens. The inspection you did not pay for at the loading point is the argument you cannot win afterwards, and on this corridor the rejected parcel has no realistic reverse gear.

What a usable Certificate of Analysis looks like for Ethiopia

• Batch or lot identification tying the certificate to the drums actually loaded, not a typical-values sheet reissued for every consignment.
• Penetration at 25 °C and softening point as measured values, each with the ASTM or EN designation printed beside the result, and each read against any narrower contractual window agreed for the shipment.
• Ductility, flash point, solubility, specific gravity and water content, each with its method.
• The ageing result in the form the tender asks for, labelled with the procedure actually run rather than described generically as loss on heating.
• For a Dega or Wurch site, an agreed low-temperature line — Fraass breaking point to EN 12593, or bending beam rheometer stiffness and m-value under AASHTO M320.
• Where a modified binder is supplied, elastic recovery, storage stability and the modification type, not merely the word polymer.
• Nothing about adhesion. Affinity between binder and aggregate cannot be certified from a binder sample alone and has to be commissioned at mix level with the project aggregate.

Who the standards body is, and why its name is part of the problem

Ethiopia’s national standardisation function has been restructured more than once. The former Quality and Standards Authority of Ethiopia was reorganised, with standards development passing to a dedicated standards body widely known as the Ethiopian Standards Agency, conformity assessment services to the Ethiopian Conformity Assessment Enterprise, and legal metrology to the National Metrology Institute of Ethiopia. The institutional arrangements and the name of the standards body have been revised again since, and this page deliberately does not assert which name and mandate are current on your shipment date. Ethiopia is a member of the African Organisation for Standardisation, and national standards are published with Ethiopian Standard designations.

That reorganisation history is not a piece of trivia. It is a practical reason to check the institution as well as the designation when a tender cites a standard, because a clause assembled from an older template may name a body that no longer exists under that name, or a designation that has been revised since the template was written. Ask the engineer to confirm the document and edition rather than assuming continuity.

Conformity assessment, described as a mechanism only

National standards bodies across the region administer conformity assessment arrangements for imported goods that fall within a declared scope. The arrangements vary in shape, but the families are recognisable and the mechanism is worth understanding whatever the current Ethiopian position is:

• The trigger is scope. A programme applies to a defined list of products, usually expressed by product category and tariff classification. Whether a given product is inside or outside that list at a given moment is the whole question, and it is the question this page will not answer.
• Verification happens either before shipment or on arrival, and which one it is changes everything. In a pre-export family the exporter or supplier applies to an inspection body appointed for the country or region of supply, the consignment is verified by documentary review of test reports, by physical inspection at the loading point and where required by sampling and laboratory testing, and a Certificate of Conformity is issued before the goods sail and travels with the documents. In a destination-inspection family the check happens after arrival, at the importer’s cost and on the importer’s clock. A supplier who assumes the wrong family has built the wrong timetable.
• There is normally more than one verification route where a pre-export programme applies: a consignment-by-consignment route, a registration route for a product whose consistency has been established, and a licensing route for a manufacturer whose quality system has been assessed. Which routes exist, and which is appropriate, is a matter for the appointed body.
• There may also be a mark. Programmes of this kind commonly sit alongside a conformity or standardisation mark applied to certain categories of goods. Where such a mark is required it is a physical marking obligation, which means it has to be established before the goods are packed rather than after.
• The consequence of getting it wrong is not a warning letter. Where a certificate is required and absent, the usual outcomes are detention, destination inspection and testing at the importer’s cost and time, penalties, or refusal of entry. On this market that penalty lands on a cargo that has already completed an ocean voyage, crossed a frontier and been hauled several hundred kilometres inland to a dry port.

Why this page does not tell you whether bitumen is in scope

Because the scope of these programmes changes. Product lists are amended, tariff lines are added and removed, appointed inspection bodies are re-tendered and replaced, and the available verification routes are revised. An undated page asserting that a product is or is not currently in scope, or naming the body currently appointed for a particular region of supply, is worse than useless: it is a statement precise enough for a buyer to plan a shipment on and wrong often enough to ruin one. So this page names no current scope and no appointed agent.

Put four questions in writing to a licensed customs broker in Ethiopia before you contract, and put them again if the shipment slips:

• Is this product, under this description and this tariff classification, within the scope of any conformity assessment or pre-shipment requirement on my intended shipment date?
• If it is, does verification take place in the country of supply or on arrival, which body is appointed, and what does the supplier have to produce and by when?
• Is any physical marking obligation attached, and does it have to be applied before the drums are packed?
• What evidence will customs expect at the point where the goods are actually entered — which on this corridor is normally an inland dry port rather than the frontier?

The customs machinery

Customs administration in Ethiopia has also been restructured: the functions of the former Ethiopian Revenues and Customs Authority were separated, with customs administration carried by the Ethiopian Customs Commission alongside the revenue ministry. Declarations are lodged electronically through the national customs system. Two structural points matter to a seller. The first is that the seller’s commercial documents are needed earlier than a first-time exporter expects, because the buyer’s bank and customs processes begin well before arrival. The second is that the goods description has to agree everywhere, which brings us to the discipline that runs through this whole page.

Petroleum bitumen falls under HS heading 2713.20. The full national subheading, and any duty, levy or tax treatment, must be confirmed with a licensed customs broker in Ethiopia. No rates of any kind are stated here.

Transit, and why the Ethiopian arrangement is not the East African one

Goods landed at the ocean gateway and destined for Ethiopia move inland under a transit arrangement to the point where they are entered, which on the Djibouti corridor is normally an inland dry port. The structural consequences are worth understanding even though the detail belongs to a broker:

• The movement runs under customs supervision with a security arrangement standing in for the charges that would fall due if the goods failed to arrive. It has a value, it has a cost, and somebody has to provide it.
• The movement is typically monitored, and the goods that arrive have to be demonstrably the goods that left. A transit consignment is not one that can be conveniently split between two receivers, decanted, reblended or partly delivered en route — which is among the strongest practical arguments against a bulk tanker on this corridor.
• There is a route and a nominated entry point, and the security is discharged only when the goods are accounted for there.
• The framework is bilateral, not regional in the East African Community sense. Kenya’s corridor traffic to Uganda and Rwanda moves under a regional single customs territory arrangement between partner states. Ethiopia is not an East African Community partner state, so nothing about that framework can be assumed to apply here. Ethiopia and Djibouti are both member states of the Common Market for Eastern and Southern Africa, and regional facilitation instruments exist — the COMESA Yellow Card, for example, is a regional third-party motor insurance scheme and is a carrier’s arrangement rather than cargo cover, and it should never be confused with cargo insurance. What actually governs your movement is a question for a broker, not an inference from a treaty name.

The constraint most supplier pages never mention: getting paid

On many markets the hard part of an export is the freight. On this one it very often is not. Ethiopia has for many years operated a managed foreign exchange regime in which the availability of hard currency, the opening of letters of credit and the sequence in which import applications were served were governed by directives of the National Bank of Ethiopia and administered through the commercial banks. Importers routinely waited in a queue for an allocation, and a perfectly agreed contract could sit unexecuted because the currency to pay for it had not been released. In 2024 Ethiopia moved to a substantially more market-determined exchange rate arrangement as part of a wider reform programme, and the rules around import payments, permits and letters of credit have been revised in consequence.

This page states no position on the current directives, the current regime or the current practice, and it states no exchange rate. What a seller needs to take from it is the shape of the risk, because it changes how an Ethiopian offer should be structured:

• Ask how the payment will actually be made before you spend time on grade and freight. Whether the buyer holds a bank permit or its current equivalent, whether the letter of credit can be opened now or is waiting on an allocation, and what the bank has told them in writing are the first questions on an Ethiopian enquiry, not the last.
• Validity periods matter more here than elsewhere. An offer priced with a short validity against a buyer whose payment mechanism has an indeterminate lead time is an offer that will have to be repriced. Say what the validity is and why.
• Insurance placement can be caught by the same machinery. Several countries in the region require marine cargo insurance on imports to be placed with a locally licensed insurer as an exchange-conservation measure. Where such a requirement applies it sits awkwardly with CIF, because CIF is the rule under which the seller procures the cover, and buyers facing it commonly contract on CFR or FOB and place the cover at home. Confirm whether it applies to your consignment with the buyer’s insurance broker before the delivery term is agreed rather than after.
• DDP is heavier here than it looks. A seller with no Ethiopian presence cannot be the importer of record and cannot access the local payment machinery, so DDP into Ethiopia should never be agreed casually.

Two further bodies a buyer may encounter

Ethiopia’s petroleum fuel import and distribution has long been organised around a state enterprise, commonly known as the Ethiopian Petroleum Supply Enterprise, and the country’s international logistics has long been organised around a state shipping and logistics enterprise which historically held the multimodal transport role for corridor imports. This page states no position on whether bitumen falls within the remit of any state import channel, licensing regime or exclusive arrangement, on what a private importer may or may not do, or on what is currently open to private operators. Those are questions for a licensed customs broker and an Ethiopian legal adviser, and the answer has changed over time. Nothing here is legal, customs, regulatory, banking or compliance advice.

Insurance, and the gap after the port

Beyond any local placement requirement, one point applies to every Ethiopian cargo: a marine cargo policy that ends at the ocean gateway does not cover eight or nine hundred kilometres of inland leg across a frontier. Inland transit cover has to be arranged deliberately and it has to run across the frontier rather than stopping at it. Carrier liability under a road carriage arrangement is limited and calculated by weight; it is not cargo insurance and it will not make a buyer whole on a full load. This is the most commonly missed line on a first corridor shipment, and on this corridor the uncovered portion is the part of the journey where the cargo is handled most.

The order to take the decisions in

Because an Ethiopian consignment has more moving parts than a coastal import, the sequence matters. Take them in this order and no decision invalidates the one before it.

• First, establish the delivery town and its altitude, and whether the goods will be entered at a dry port or elsewhere. Everything else follows from this and nothing can be settled before it.
• Second, establish how the payment mechanism will work and what the buyer’s bank has confirmed in writing, because on this market that is the constraint most likely to stop the transaction.
• Third, take your own legal and compliance advice covering the goods, the parties, each customs territory on the route and the payment mechanism, from advisers accountable for the opinion.
• Fourth, put the conformity question to a licensed customs broker in Ethiopia, in writing and dated, early enough for any pre-shipment verification or marking to be arranged in the country of supply.
• Fifth, settle the corridor and the current operating status with a freight forwarder, in writing, including which gateway, which crossing and which mode they can actually execute for this commodity, and whether the movement is multimodal or unimodal.
• Sixth, choose the packing from the length of the land leg and the equipment at the receiving yard, then convert tonnage into vehicles with the forwarder rather than with a calculator.
• Seventh, read the tender’s binder clause, settle the designation, and fix where in the band the batch must sit — toward the softer half for a plateau site, the harder half for a lowland one — and add the low-temperature or adhesion lines the site needs and the export sheet does not carry.
• Eighth, choose the Incoterms 2020 rule from the mode and name the place precisely, and settle in the same clause who arranges the frontier formalities and who bears the cost of a vehicle standing and waiting.
• Ninth, write the test schedule into the contract and inspect at loading, with sampling to ASTM D140, sealed retained samples held by both parties, and a recorded drum count, drum condition and seal number at every handover from the load port to the dry port.