By: Prof. Adriana Ana-Maria Davidescu, Bucharest University of Economic Studies, Romania
“A business school is at its most useful when it stops describing a problem and starts measuring it. This is what that looks like.” — Vasile Alecsandru Strat, PhD, Dean, Bucharest Business School
A corridor that became real faster than the plans for it
For most of the last decade the Trans-Caspian route was discussed as a theoretical alternative — a line on a map that existed mainly in strategy documents. It moved 0.8 million tonnes in 2019. In 2024 it moved 4.1 million. Somewhere in between, the conversation stopped being about whether the corridor would work and became about who would serve it.
Romania has an unusually strong hand. Constanța is the largest port on the Black Sea and the terminus of an EU corridor, and its historic rival has effectively been removed from the market by the war. Market signals have followed: a stake taken by AD Ports Group, a memorandum on a Caspian–Constanța link.
What has been missing is not ambition. It is a number. Before a port authority, a ministry or an operator commits capital, somebody has to answer a narrow question: if one condition changes, how much cargo actually moves, and is it worth what it costs? That is the question our team set out to answer.
The answer came from a gravity model of Romania’s seaborne trade — the standard tool for explaining why cargo moves between two places — estimated on eight years of bilateral flows with forty-five partners, measured in dollars and in tonnes at the same time. Measuring both matters more than it sounds. Romania’s seaborne trade has grown in dollars and shrunk in tonnes, and a port that watches one series without the other will misread what it is actually handling.
Everything depends on one leg
The answer turned out to be more concentrated than we expected.
Half of Romania’s seaborne trade by value is already with four corridor partners — Turkey, Kazakhstan, Azerbaijan and Georgia. But those partners are not in the same condition. Measured against what comparable, well-served relationships achieve, Turkey and Italy operate in the mid-nineties. Kazakhstan, Azerbaijan and Georgia operate at 57%.
The three of them reach Constanța through the same maritime leg, via Poti and Batumi. They are not three separate underperforming relationships. They are one underserved link, counted three times.
That changes what the problem is. Unrealised trade spread thinly across thirty partners is a condition to live with. Unrealised trade sitting on one leg is something a commercial decision can address.
What a ship a week is worth
We priced the available measures on the same model, and the ranking was not close.
A regular ro-ro service between Constanța and Poti or Batumi, with the terminal capacity to serve it, is worth more than every procedural reform combined — on the order of seven billion dollars of additional seaborne trade. Faster customs release, a single corridor tariff and consignment note, and digitalisation together come to roughly a third of that. Those three are not speculative: the European rules that carry them, electronic freight transport information in 2027 and the EU Customs Data Hub in 2028, already have dates attached.
This is not an argument against the procedural measures. They cost almost nothing, they can start now, and the case for them is avoided loss rather than upside: on current trends, doing nothing through 2028 quietly costs around two and a third billion dollars of trade, roughly three times what the cheapest package would have delivered. But a ministerial order is not a substitute for a ship. Two of the three best measures by cost-effectiveness are coordination problems, not capital problems — which is a comfortable finding right up to the point where it is used as a reason to postpone the one that is.
The number nobody in the room could give us
Here is the part that should interest anyone operating on this route.
Acting on the Georgian link alone outperforms a general push to improve connectivity everywhere — but only above a threshold. If more than roughly 43% of Caspian cargo moves through Georgia, the targeted strategy wins by a wide margin. Below that, it does not. And much Kazakh crude does not go that way at all: it travels as CPC Blend out of Novorossiysk, or via Supsa, and a ferry at Poti does nothing for those tonnes.
We cannot observe that share from trade statistics. Operators can. This is why the model was taken to a panel of corridor practitioners at our Shipping & Port Leadership Summer School in Constanța rather than published as a conclusion, and why it travels with a questionnaire attached. The questionnaire takes ten minutes and is open to anyone working the route. The division of labour is straightforward: the model shows practitioners where the potential sits, and practitioners supply what the model cannot see.
It also means the headline figure is an upper bound with a condition attached, which is a less satisfying thing to publish and a more useful thing to act on.
A window, not a trend
One last observation, and it is the one most easily missed.
Romania’s seaborne trade has grown substantially in dollars since 2017. Almost all of that growth is price. In physical terms the port handles slightly less cargo than it did, and in real terms trade peaked in 2021. Even the most favourable path we model leaves 2027 below that peak.
So the corridor question for Constanța is not expansion. It is recovery, with a time limit. The combination that makes the next two years unusual — Odesa out of the market, Caspian capacity being built, carriers still deciding where to put their calls — will not hold indefinitely. A terminal decision and a commercial agreement take two to four years to produce a sailing. Counted backwards from the moment the window closes, the decision is not a 2028 decision.
The corridor gap is not a mystery to be explained. It is a lever, it is attached to one link, and we now have a reasonable idea of how hard it would have to be pulled. Whether anyone pulls it in time is not a modelling question.
