Canada Puts C$254m Into Saint John Corridor, Aiming for 850,000 to 1,000,000 TEU a Year

The Analysis

Spread the map on the table. Put a finger in the Bay of Fundy, on the New Brunswick side, and Saint John is sitting there, in the one place on Canada's east coast where the water rises and falls further than it does anywhere else on earth. It has always been a port. For most of a century it was a port that things left from: pulp, paper, potash, crude. Containers were the side business.

On 21 October Transport Canada committed up to C$254m to the corridor behind that port. It is the first draw on a C$5bn Trade Diversification Corridor Fund and the first project that fund has ever landed. New Brunswick put in C$116.5m and the project's other partners C$92.5m, which makes the announced envelope C$463m. Add the C$76m that came earlier from the National Trade Corridors Fund and the public money stacked on this one corridor reaches C$539m. The figure travelling with the announcement is 850,000 to 1,000,000 TEU a year.

That last number is the one I want to take apart, because it is not what this money is building. It is what the berths were already built to handle before a dollar of the C$463m was announced: DP World's West Side rebuild had already taken Port Saint John's design capacity from 150,000 TEU to as much as 1,000,000 TEU. The cranes were the private half. What the public money buys is everything a ship has to pass through before it reaches a crane and after it leaves one.

Money like this does not move because a port asks for it. It moves because something upstream pushed it. Two things pushed it here. One is commercial: the West Side berths exist and they are not full, and a terminal operator that has built a million TEU of quay does not sit quietly while the rest of the corridor catches up. The other is political, and it is the one that explains the name of the fund the money came out of. A Trade Diversification Corridor Fund is not a name anybody invents in a calm decade. It exists because Ottawa spent two years watching its largest customer put tariffs on Canadian goods and finished that period believing a country that sends most of what it exports to a single neighbour needs a second way out. C$5bn was the answer to that belief. Saint John is the first place the answer got spent.

What makes it first rather than fifth is geometry. Saint John is the easternmost deepwater port in Canada with rail directly behind it, and it is the closest Atlantic port to the St. Lawrence. A box landed there sits as near to Quebec and to Ontario as an Atlantic port can put it without going up the river. That has always been true and it was never enough, because the weak half of the geometry was the land side. The railway into the port runs through one yard, at Bayshore, and that yard is the port's only rail access. Reinforcing its shoreline is not a finishing touch. It is the difference between a container leaving the terminal on steel and leaving it on a truck, and a port that cannot promise the steel cannot be sold as a rail port.

Read the list of works and the argument inside it is hard to miss. Dredging, to widen the high-tide window for ships arriving and leaving fully laden by at least 30 per cent. A hundred and three kilometres of track relaid as continuous welded rail, with the mainline lifted to 64 km/h. A 2,787 m² inspection building for the Canada Border Services Agency. Shoreline reinforcement at Bayshore. A road-over-rail separation at Simms Corner. Every item is a link or a gate. Not one of them is a berth. The berths are there already, paid for on the private side, and every dollar of the public C$463m goes into the parts of the system that only became the constraint after the berths stopped being it.

Ten years ago the page would have looked different. A Canadian importer bringing Asian cargo into Montreal had two real options and one expensive third: rail it the width of the country from Vancouver, or land it on a US East Coast port and pay for the border on the way north. Saint John was what you used when the first two were jammed. It was a relief valve with a cruise terminal attached to it. The claim buried in the C$463m is that a relief valve with its own deep-draft window, its own welded rail and its own examination capacity is not a relief valve at all, it is a fourth door. Whether it becomes one is decided at gates the announcement puts into a subordinate clause.

Here is the thing I have not found written down in any of the coverage, and it is the reason I would not let anybody book against a million TEU. A corridor is not a terminal. It is a set of gates in series, and what it passes in a year is what the narrowest gate passes, not what the widest one can. If Port Saint John hits 1,000,000 TEU, that will be a fact about cranes. For the corridor to move 1,000,000 TEU, four other things have to move it too, and the smallest of them is named once, in one clause, as a 2,787 m² building.

So put the assumptions on the table and work it. Take the top of the target: 1,000,000 TEU a year. Assume 55 per cent of that is laden import, which is 550,000 boxes. Assume 4 per cent of import boxes get opened and examined, which is 22,000 examinations a year. Assume the building runs 250 days a year on two eight-hour shifts, so 4,000 productive hours. Assume a box occupies a bay for three hours between being set down and being lifted off again. Assume each working bay needs 105 m², which is the 30 m² a forty-footer stands on plus the aisle, the door swing and the room to walk around it: 2,787 divided by 105 is roughly 26 bays. One bay on 4,000 hours turns 1,333 examinations a year, and twenty-six bays turn a little under 34,700. Against 22,000 the building is running at about 63 per cent of its ceiling. Not full. But the distance from not full to full is shorter than it looks.

The distance is the whole point. Twenty-two thousand sits at 63 per cent, and the building chokes at an examination rate of about 6.3 per cent of import boxes, which is under two percentage points above the rate I assumed. It also chokes if three hours of bay occupancy becomes four and three-quarters, which is what happens the moment the same concrete has to hold boxes waiting on a laboratory result, or on a broker returning a call, or on a truck that has not turned up. When it chokes, the queue does not form at the crane. It forms inside the building, and the million TEU stays true of the berth while the corridor delivers whatever 2,787 m² lets through.

The dredging carries the same shape of caveat, and its number is stranger than it reads. The announcement promises at least 30 per cent more high-tide window for laden deep-draft ships. It does not promise more ship. It promises more clock. Assume the usable window at each high water was three hours, and assume two high waters a day, which is 706 a year. Three hours becomes three hours and fifty-four minutes, and across the year 2,118 usable hours become 2,753. Six hundred and thirty-five hours appear from nowhere. What they buy is not a bigger ship. What they buy is the hour in which a late arrival can still get in and the hour in which a late departure can still get out. Note that 30 per cent is a floor in the wording; the dredging contract may well deliver more, and I would not put that extra into a plan until somebody has surveyed the channel after the work.

For the people who decide where a ship calls, that hour is worth more than depth. A carrier does not publish a call because a channel is deep on average. It publishes a call because it believes a berth window will hold on the same day of the week, and it will not believe that on the strength of a dredging contract. Civil works can hand a port a longer window. Only a timetable can hand it a reliable one, and the timetable belongs to an office that has three other Canadian ports on the same map.

Who gains from this depends on which way your boxes travel. For an exporter in New Brunswick or eastern Quebec, the 64 km/h matters more than the dredging: 103 km at 64 km/h is an hour and thirty-seven minutes of running time, and every minute cut out of that is a minute of reefer fuel or a minute of transit that stops being paid for. For an importer routing Asian cargo to Montreal through a US East Coast port today, the proposition is a straight bet that the border crossing at the far end costs more than the extra steaming time. For anybody whose cargo ends up in the American Midwest, none of it applies, because Saint John is a long way from Chicago and geography does not negotiate. For Halifax and Montreal the corridor is competition rather than relief, and both of them should be watching the C$4.75bn still sitting in the fund far more closely than the C$254m that has come out of it.

Sixty-four kilometres an hour is worth reading as a number about the corridor's far end rather than about the port. It is the mainline target across 103 km, and a target is not a timetable: on any line with curves and grades the average over the whole stretch sits under the figure, and in a February the average sits under it again. What the relay really changes is the spread. A corridor that used to promise a transit somewhere inside a wide band will afterwards promise one inside a narrower band, and for anyone holding a fixed weekly delivery appointment into Quebec that narrowing is worth more than the hour and thirty-seven minutes it was carved out of. It also tells you where to look for the next bottleneck, because a corridor that reliably delivers a box in a narrow band will hand its problem to whatever sits at the far end of the 103 km, and that has not been rebuilt.

That C$254m is the least interesting number in the release. It is a little over 5 per cent of the C$5bn fund it was drawn from, which leaves C$4.75bn unallocated and makes this project the template the rest will be spent against. The template is worth reading closely, because it is five lines long: deepen the approach, relay the rail, build the examination building, reinforce the one yard everything depends on, take the road off the track. Every one of those five lines has a twin at Halifax, at Sept-Îles, on the Roberts Bank causeway and on the rail between Montreal and the Atlantic. If you want to know which of them gets the next cheque, the announcements are the wrong place to look. Engineering tenders will tell you the shape of the work a year or two before a press release tells you its name.

The dates are the part to hold loosely. The Premier's office has construction possibly starting in spring 2027 and finishing in 2029 or 2030. Possibly is carrying weight in that sentence, and so is up to in the federal figure. Corridor money is paid against milestones, which means the cash arrives when the work does, and if the first shovel goes in during 2029 the volume does not turn up in 2030 either. Use the late end of both ranges and nothing that follows will surprise you.

There is a second timing effect that will reach you earlier than any capacity does, and it points the wrong way. A hundred and three kilometres of track does not get relaid on a railway that has been closed. It gets relaid on a railway that is running, in windows, in the months the network is busiest. For a construction season or two the corridor will move its freight through a worksite. If you ship on that line, the question for 2027 and 2028 is not when the extra capacity lands. It is how many hours a day the line is single-tracking, and what your transit looks like in a week when it is.

Nobody holds a press conference for the empties, and the empties decide whether this works. A port that lands 550,000 laden import boxes has to find homes for the metal afterwards, and on a corridor this new the nearest place that wants it is a long way off. If the empties leave empty, every one of them takes a slot on the 103 km and a slot at the berth that a loaded box could have used. If they leave loaded, the corridor has an export base behind it and the whole thing reads differently.

The argument for writing the corridor down is short and it deserves saying out loud. A million TEU needs a carrier to bring a deep-draft mainline call, and a carrier only does that when the cargo is there before the berth is. Ports have been built on that circle before and some of them are still standing empty inside it. If tariff pressure on Canadian exports eases, the diversification case softens and the will to finish on time goes with it. If the volume comes anyway, the constraint moves straight past Saint John to whatever the 103 km joins onto, and the next round of money has to be spent inland.

Ömer Kaya would put the colder version of that question, which is whether the lane is worth running at all once the trucks at both ends have been counted; on the evidence of the works list, the people who designed this envelope have counted them, which is why a grade separation at Simms Corner and a shoreline at Bayshore sit in the same document as the dredging.

Set the money against the volume and the unit cost is readable. The corridor's annual capacity goes from 150,000 TEU to a range topping out at 1,000,000, so the addition is 850,000 annual TEU, and the crane half of that sits on the private side of the accounts. Against it, C$463m of announced public money is C$545 per annual TEU, and with the earlier C$76m included the C$539m works out at C$634. Twenty years of asset life at no discounting makes the second figure about C$32 a TEU. A shipper moving 1,000 TEU a year through the corridor is moving it on top of roughly C$634,000 of public capital. Nobody sends a bill for that. It is the reason the corridor will be priced to pull volume in the year it opens, and the reason the first two years after it opens are the years worth negotiating in.

What a shipper should do with all of this in 2026 is cheap and takes an afternoon. Find out where your boxes are physically examined on the corridor you use now, and how many square metres that building is, because that building is the gate you will be queuing at in 2030 and it is standing there today. Treat 2027 and 2028 as the slow years on the Saint John line rather than the fast ones, and have a truck number ready for the weeks the track is being relaid. Work out which of your lanes would shift if an Atlantic port with 64 km/h behind it started publishing a fixed-day call, and what the shift would do to your inland leg, because that answer tells you whether the corridor is an opportunity for you or a rival to the port you use now.

The alternative, which is to route through Halifax or Montreal today and look again in 2029, is right for most cargo and wrong for two kinds. It is wrong for anything moving within 300 km of the corridor, where the inland leg is most of the cost and always has been. It is wrong for anything that will be competing for room on the corridor in 2030, because by then the sheds near Bayshore will not be cheap and the carriers will have chosen their customers. The trap is reading 1,000,000 TEU as a floor. It is the top of a range whose bottom is 850,000, and the range was published before the examination building had been sized against it.

Ports are the slowest objects in a supply chain and the reasons one works are always dull. A hundred and three kilometres of welded rail. Two thousand seven hundred and eighty-seven square metres. Three hours and fifty-four minutes of tide. That is what a fourth door is built out of, four years from now, in the place where the water moves further than it moves anywhere else on earth. The map on the table will look the same in 2027. It may look different in 2030, and if it does, the C$4.75bn still in the fund will be the reason for it rather than the C$254m that came out.

↑ Back to the key points

— By Tomás Ferreira