The Unseen Queue
Picture a mid-size container terminal handling 1.2 million twenty-foot equivalent units a year. Planners anticipate a 30% rise in traffic, expand the facility accordingly, and then watch, with some bewilderment, as average yard dwell time climbs from 3.2 to 5.8 days within four years. Trucks queue for two to four hours per visit despite the new capacity. This is not an unusual story at major ports, and it keeps happening for the same reason: the way planners count congestion costs is structurally incomplete.
The Induced Demand Problem
The mechanism is familiar from highway economics. Added capacity reduces short-term friction, which in turn releases suppressed demand. Shippers reroute. New traffic fills the gap. Port economists have a name for the broader dynamic (the Downs-Thomson paradox), but the traffic models used in most feasibility studies are static. They do not capture how lower access costs change shipper behaviour over time. There is also a subtler problem: terminal operators, shipping lines, and port authorities all have institutional reasons to present optimistic demand forecasts. The projections, in other words, are not always disinterested.
The historical record on this point is not encouraging. A review of major port expansions completed between 2000 and 2015 across North America and Northern Europe found that actual throughput in the five years following completion exceeded feasibility-study projections in roughly two-thirds of cases, not because the ports performed better than expected, but because demand growth outstripped the capacity that had just been added. The expansions had, in effect, created the very congestion they were designed to relieve, simply on a larger scale and at greater public expense.
The Hidden Costs of Congestion
When planners do account for congestion, they tend to focus on the legible items: vessel wait times, gate delays, berth utilisation rates. Reasonable enough, as far as it goes. But those costs are largely passed downstream to importers and, eventually, to consumers, which makes them easy to treat as someone else's problem. The costs that go genuinely uncounted tend to sit outside the terminal fence entirely. Road network congestion generated by port traffic can translate into millions of dollars in additional labour costs for trucking firms, none of which appears in the port authority's spreadsheet. Inventory costs are similarly invisible: when containers sit in the yard longer than planned, importers must finance more stock in transit, tying up working capital that would otherwise be deployed elsewhere. Small per-unit figures, multiplied across thousands of shipments, become significant sums.
There are environmental costs in the same category. Idling trucks and vessels waiting at anchorage produce emissions that are rarely costed into expansion appraisals, even in jurisdictions with active carbon pricing regimes. The communities nearest to port approaches, typically lower-income residential areas with limited political influence, bear a disproportionate share of the resulting air quality burden. That distributional dimension seldom surfaces in the investment case presented to a port authority's board, and almost never in the press releases that accompany a groundbreaking ceremony.
The Measurement Gap
Most ports do not use dynamic pricing to reflect real-time congestion levels. That absence matters more than it might seem. Without a price signal, importers have no reason to adjust the timing or volume of their shipments during peak periods, so the yard stays full and the queues persist. Compounding this, port authorities typically report raw throughput figures rather than throughput efficiency, which makes it genuinely difficult for outside observers, or indeed for the port itself, to assess how congestion is eroding the value of installed capacity. A terminal moving more boxes than last year can still be performing worse, if each box is taking substantially longer to clear.
The data problem runs deeper than reporting conventions. Many ports lack the integrated systems needed to track a container's full journey from vessel discharge through yard storage to gate exit in a single, queryable record. Where that data exists, it is often treated as commercially sensitive and withheld from regulators and researchers. The result is that the academic literature on port congestion costs relies heavily on a small number of cases where data happened to be available, which may not be representative of the industry as a whole. Policymakers are, in effect, making billion-dollar infrastructure decisions on the basis of an incomplete and potentially skewed evidence base.
A More Honest Accounting
A congestion cost model worth trusting would need to do at least three things: incorporate induced demand rather than assuming traffic is fixed, extend the cost boundary beyond the terminal gate to include hinterland network effects, and stress-test forecasts against scenarios with materially higher demand than the central case. Some European port authorities have moved in this direction, adopting broader social cost accounting frameworks that fold hinterland costs into their investment reviews. The Port of Rotterdam's long-term planning process, for instance, has for some years included hinterland accessibility modelling as a formal input to capacity decisions, a practice that has influenced both the sequencing of investments and the conditions attached to terminal operator licences.
The practice is not yet standard, and whether it will become so depends partly on political will and partly on who is asking the questions. Expansion projects generate construction contracts, long-term lease revenues, and employment announcements, all of which create constituencies with an interest in approval rather than rigorous scrutiny. Anyone examining an expansion proposal, whether as a regulator, a creditor, or a neighbouring municipality, would do well to start by asking whose costs are inside the model and whose are quietly left out. The answer to that question tends to determine not just the economics of the project, but who will be living with its consequences long after the ribbon has been cut.