Savannah Expands Garden City Terminal Capacity With New Berth

The Analysis

Spread a map of the North American Atlantic seaboard across a table and run a finger from the Chesapeake down to the Florida line. For most of a century that coastline looked settled: geography decided where the harbours were, the railroads decided which of them mattered, and the hierarchy barely moved from one generation to the next. Ten years ago this coast read differently. Savannah was not the port that came first in a routing meeting; it was the alternative somebody raised when the obvious ones were congested, the second call you made after the first one failed. On 6 May 2026 the Georgia Ports Authority opened an additional berth at Garden City Terminal, lifting annual container capacity by roughly 1.5 million TEUs, and it said the expansion supports growth across the Southeast and reduces vessel queue times. It also flagged rail connectivity upgrades as the next priority to keep inland flow smooth. That second sentence is the one I want to talk about.

A berth is not news measured in weeks. Concrete poured on a tidal river is a twenty to thirty year promise; the cranes that will stand on it are specified for vessels that have not been ordered yet; the channel that feeds it was dredged on the assumption that the ships will keep growing. When a port authority builds a berth it is not reacting to a quarter. It is placing a long public bet on where the cargo of the next decade wants to go, and once the bet is poured, it cannot be moved. That is why a berth deserves the ten-year lens and not the weekly one.

What actually changes here is not the length of quay. It is the location of the constraint. Every container network has one point that sets the pace for everything behind it, and ports spend their whole lives pushing that point around. Add waterside capacity and, if the landside cannot absorb it, the constraint simply walks inland and reappears as yard density, chassis supply, rail car availability and appointment slots. The authority has told us, in its own words, that rail is the next job. I take that at face value. I also think it is a warning wearing the costume of a schedule.

Let me set out the hard facts first, because everything else hangs off them. The Georgia Ports Authority opened an additional berth at Garden City Terminal in Savannah. Annual container capacity rises by roughly 1.5 million TEUs. The authority says the expansion supports Southeast US growth and reduces vessel queue times. It flags rail connectivity upgrades as the next priority to keep inland flow smooth. That is the whole of what was published, and I am not going to add to it. I will not quote a throughput figure, a rank among US ports, a dollar cost for the project or a tonnage number, because none of those were in the release and none of them belong in a sentence I wrote. If you need them, go to gaports.com and read the announcement yourself. What I can do with four published facts is more useful than what most people do with twenty invented ones.

Now take the ten-year view, because a berth cannot be understood any other way. The Southeast has been pulling population, distribution centres and manufacturing towards itself for a long stretch of years, and cargo follows the warehouses, not the other way round. A container port does not create that demand; it either keeps up with it or it becomes the reason the demand goes somewhere else. Look at the map again and the logic is almost crude: the Southeast sits close to a very large share of the American consumer, it has land for the big boxes that modern distribution wants, and its highway and rail bones reach deep into the interior. A port that adds berth capacity in that position is not gambling on a trend. It is refusing to lose to one.

The network question is the one I care about, and it goes like this: which bone in this network just moved? Not the quay. The quay was the constraint that everybody could see, the queue of ships waiting for a slot, the picture that ends up in the newspaper. A new berth attacks exactly that. But a terminal is a system with four throats in series: the berth, the yard, the gate, and the inland connection beyond it. Open one throat and the flow immediately tests the next one. Since the authority itself names rail as the next priority, the honest reading is that the landside is now the throat under strain. Everything downstream of the new concrete, from the stack to the ramp, is where the next five years of friction will be felt.

Let me put a number on that, using figures from the release and assumptions I will state plainly. Assume that roughly 60% of the added 1.5 million TEUs eventually moves inland by rail rather than by truck. That proportion is my assumption; the announcement does not say it. Then rail has to absorb about 900,000 additional TEUs a year. Divide by 365 and that is roughly 2,466 TEUs a day, and divide again by 24 and it is about 103 TEUs an hour, every hour, including the night and including Christmas morning. That is the real meaning of a berth: somewhere inland, a railroad and a set of ramps now has to swallow an extra hundred containers an hour, forever, or the yard fills up and the new berth stops being an asset and becomes a very expensive parking lot.

Now shrink it to the size of one importer, because that is the scale at which you actually make decisions. Assume you move 40 forty-foot containers a month through Savannah, which is 80 TEUs and 960 TEUs a year. The added annual capacity of 1.5 million TEUs is then equivalent to roughly 1,560 businesses the size of yours. That ratio is the point: none of this capacity was built for you, and none of it waits for you. Your leverage is not that you matter to the port. Your leverage is that you noticed early and arranged yourself accordingly, which is a different and much more reliable kind of advantage.

The part of the announcement that touches your week is the queue. The authority says the berth reduces vessel queue times. Assume, as an illustration and not as a published figure, that your average berth wait on this service falls by 10 hours. Ten hours sounds like a gift, and it is, but only if it converts into something you can use. It converts if your inland side can take the box when it lands. It evaporates if the box arrives ten hours earlier and then sits because the rail cut-off was missed, or because there was no chassis, or because the appointment system had no slot until Thursday. Time saved at the berth is only worth the hours it does not lose again in the yard. So the right question to ask your forwarder is not how much faster the ship gets in. It is how many hours of free time you actually get, and what the terminal does with a box that outstays it.

Here is the arithmetic I would run on that second question. Assume your free time at the terminal is 96 hours, which is four days, and assume a landside snarl adds 24 hours of dwell before your box is gated out. Twenty-four hours out of 96 is a quarter of your buffer gone, and it is gone before a single mile of inland transport has happened. Assume you run 40 boxes a month, so that is 960 box-hours a month of exposure created by a delay that never appeared on the vessel schedule. Now assume the pattern is not one bad week but a permanent 12-hour degradation in inland fluidity, which is what happens when waterside capacity runs ahead of rail for a couple of years. Twelve hours out of 96 is 12.5% of your buffer, permanently, on every import, which is the kind of slow leak that never shows up as an incident and always shows up as a cost line.

So what do you actually do, and who does it. Within one week of 6 May, I want your logistics manager to produce a single page for Savannah: contracted free time in hours at the terminal, the current rail cut-off in hours before vessel arrival, the chassis arrangement in force, and the appointment lead time in days. Four numbers, one page, one owner, and a review date. If any one of those four cannot be answered, that blank is your risk register entry for this quarter. By the end of May, that page goes to whoever signs the inland contract, with one specific instruction: negotiate free time and appointment access in hours, not in days, and write the failure remedy into the contract rather than leaving it to goodwill.

The second move is a calendar move, and it is cheap. Assume the rail upgrades the authority flagged take several years to design, fund and build, because things built on that scale always do. Between now and then, the landside runs ahead of, or behind, the waterside in cycles, and every cycle has a bad season. So build yourself one alternative routing and keep it warm rather than cold. Assume a second gateway on the same coast, or an inland ramp that bypasses the terminal gate entirely, and assume you test it with real volume once a quarter instead of discovering it in an emergency. A test costs you a few containers a quarter. An emergency migration costs you the bad season, which on a 960-TEU-a-year importer is eighty boxes of chaos.

The third move is inventory, and it is the one that decides whether any of this ever hurts you. Assume your inland leg is now the variable one rather than the ocean leg. Then the buffer belongs in the inland schedule, not on the water. If you carry safety stock sized against total lead time, recalculate it with the inland variance separated out, and set the trigger for a review at the point where the inland share of total lead time exceeds a third. I am not going to hand you a universal number of days, because it depends entirely on your product and your customer promise. What I will say is that a safety stock model calibrated in an era when the berth was the bottleneck will systematically understate the risk in an era when the ramp is.

Alternatives deserve their costs stated, because every one of them takes something. Routing through a different port on the same coast buys you a different congestion profile and costs you the inland mileage you already optimised; if your distribution centres sit in the Southeast, that mileage is not trivial. Moving boxes to an inland ramp early buys predictability and costs you terminal storage you would otherwise have used for free. Shifting to a second gateway permanently buys resilience and costs you the relationship depth you have built with the people who know your cargo. Splitting volume across two ports buys you an option and costs you the focus that gets you noticed when capacity is tight. None of these is wrong. All of them are paid for, and the price is always measured in hours somewhere else in your chain.

Every port I have worked on is really a fight about hinterland rather than quay. The quay is where the fight is visible; the hinterland is where it is decided. Ten years ago a routing decision in this part of the world was made mostly on ocean transit and terminal productivity, and the inland leg was treated as a fixed cost that everybody simply paid. What has changed across a decade is that the inland leg stopped being fixed: distribution centres moved, the mileage behind each gateway changed, and the railroads rebuilt their own networks around intermodal rather than around carload. A port that understands this builds berths and then immediately starts talking about rail, which is exactly what the authority has done here. A port that does not understand it builds berths and then wonders why the yard is full. This announcement contains both halves, and the order in which they were said matters.

There is a second piece of arithmetic worth doing, because ship size turns a smooth flow into a pulse. Assume a large vessel discharges 3,000 TEUs during an 18-hour port call; that is my assumption, not a published figure. Then the yard has to absorb roughly 167 TEUs an hour while the cranes are working, against the roughly 103 TEUs an hour of additional rail throughput I calculated above. Assume both figures are roughly right and the rail side runs short by about 64 TEUs an hour during the discharge window.

That gap does not disappear. It is stored, and the storage medium is the stack. This is why a bigger ship does not simply mean a cheaper box. It means a sharper peak, and every sharper peak is paid for in yard space and in the hours a box spends waiting for the next rail cut. Ask your terminal contact how many rail cuts a day your boxes can realistically make. That answer tells you more than any capacity headline ever will.

Then look at where your own boxes actually go, because the ports on this coast compete over shared inland territory rather than over separate fiefdoms. Assume your distribution centre sits about 250 miles inland from the terminal, an assumption for the sake of the arithmetic. Ten hours saved at the berth is worth very little if it is then spent waiting for a truck appointment, and it is worth a great deal if it converts into a same-day gate-out. Run the ratio: ten hours against an inland leg which, at that distance, is a matter of hours rather than days.

If the inland leg is six hours of driving plus an appointment, then the saving at the berth is the same order of magnitude as your entire inland transit, and losing it again to dwell means you handed back a whole leg. That is the arithmetic that should decide whether you chase the faster berth or the better appointment slot. I know which one I would chase, and it is not the one in the press release.

Is this a pulse or a turn? I get asked that about every port project and I answer it the same way: you cannot tell in the first year, and you can usually tell by the fifth. A pulse is capacity poured into a cycle that then reverts; a turn is capacity poured into a shift that keeps going. The two observable signals are unglamorous and they are free. Count how many services call, and count what is being built inland within a day's drive of the terminal. If the call count holds while the distribution sheds keep rising, the berth was a turn and you should commit to it. If the call count slips while the sheds keep rising, the cargo is arriving through somebody else's gate, and the berth was an insurance policy that never had to pay out. Neither answer needs a consultant. Both need somebody from your company to drive out there once a year and look with their own eyes.

The traps are mostly about the words people use. Queue time is a vessel-side measure and it tells you nothing about how long your box sits once it is ashore. Capacity is a waterside figure and it does not include the yard, the gate or the rail. Priority, when an authority uses it about a project, means the intention exists, not that the money is spent or the steel is ordered, and the gap between those three states is measured in years. And smooth, as in keeping inland flow smooth, is a goal and not a commitment; nobody owes you a remedy because a press release used an adjective. Read every port announcement as a statement about waterside intent until proven otherwise, and then go and measure the landside yourself.

I should say what would change my reading, because a view with no exit condition is just an opinion with a louder voice. My argument assumes the Southeast keeps growing and the rail upgrades arrive late relative to the berth. Assume instead that the rail work lands quickly and the inland side runs ahead of demand for a few years. Then the new capacity genuinely shortens door-to-door time, the yard stays fluid, and the friction I am warning about never materialises; in that world the berth is simply good news and the right response is to concentrate more volume here rather than diversify it.

Assume, at the other extreme, that the growth the expansion was built for does not show up, and you get the opposite problem: empty stacks, a port with capacity to sell, and a market where the lever is back in your hands for a while. Watch two things to decide which world you are in: whether the rail projects move from announcement to construction, and whether the yard looks full on an ordinary Tuesday. Both are visible from the road.

Ten years from now this berth will be a detail in a much larger picture, and the picture will be about which coasts grew and which corridors carried the growth. The map I spread on the table at the beginning will have been redrawn slightly, as it always is, by concrete and by rail and by the slow movement of where people live. What I would like you to take from 6 May is not a fact about Savannah. It is a habit: when a port tells you it has added capacity, go and look at what sits behind the berth, because that is where your hours are going to be won or lost, quietly, for the next decade.

↑ Back to the key points

— By Tomás Ferreira