HMM Commits $1.61bn to 12 Boxships, Adding Eight 13,000-TEU Newbuilds

The Analysis

HMM told the market on 15 October that it is spending US$1.61bn on 12 containerships. Eight are 13,000-TEU newbuilds it will order. Four are resale newbuildings already on the ways: three 2,800-TEU feeders and one 1,700-TEU ship. Everything is due by December 2029.

Every number in that sentence deserves a second look, and we are going to give it one. The total is the least informative figure in the package. The split, the size, the delivery ceiling and the lines HMM left out say far more.

There is also one piece of arithmetic missing from the disclosure, and it is the piece that decides whether this story means anything to an importer or an exporter before 2029. Alphaliner has HMM at 97 boxships and roughly 1m TEU, with 36 ships on order. Twelve on top of that is an orderbook of 48, half the size of the operating fleet. Twelve inside that 36 is an increase of nothing at all. HMM did not say which it is.

Start with what the announcement actually contains. Twelve ships, US$1.61bn, eight 13,000-TEU newbuilds to be ordered, four resale newbuildings to be bought, all delivered by December 2029. That is the whole of it. No yard for the eight, no seller for the four, no ship-by-ship delivery schedule, no split between owned and chartered tonnage, and no propulsion.

The arithmetic you can do anyway is not small. Eight 13,000-TEU ships is 104,000 TEU. Three 2,800-TEU feeders is 8,400. The 1,700-TEU ship is 1,700. The package therefore carries 114,100 TEU. US$1.61bn over twelve hulls is US$134.2m a ship. US$1.61bn over 114,100 slots is US$14,109 a TEU.

Both of those averages hide the thing worth knowing. The eight big ships are 66.7% of the hull count and 91.1% of the TEU. The four small ones are 33.3% of the hull count and 8.9% of the TEU. A single blended figure spread across those two groups is not a price. It is a smear.

So run it backwards. The assumptions below are ours, not HMM's, and they exist because HMM did not break the cheque down. Put the eight 13,000-TEU hulls at US$140m each and they consume US$1.12bn, leaving US$490m for the four resales, which is US$48,515 a TEU. Put them at US$150m and the resales cost US$410m, or US$40,594 a TEU. Put them at US$160m and the resales cost US$330m, or US$32,673 a TEU. Across that whole range the eight newbuilds land between US$10,769 and US$12,308 a TEU, and the four resales land between US$32,673 and US$48,515 a TEU. The secondhand tonnage costs between 2.7 and 4.5 times as much per slot.

Nothing in that calculation is a fact HMM handed us. All of it is a stress test of whether US$1.61bn holds together internally, and it does, under several different readings. What it cannot be is a reading in which the four resales came cheap. Every plausible split has HMM paying between US$231m and US$366m above newbuild economics for 10,100 TEU of feeder capacity. That is 14% to 23% of the package spent on the calendar rather than on steel.

One more check on the fleet it joins. Alphaliner's 97 boxships at roughly 1m TEU works out to about 10,300 TEU per hull on average. The eight newbuilds at 13,000 sit above that average and the four resales at 2,800 and 1,700 sit far below it. HMM is not upsizing its fleet, it is stretching it at both ends at once, and the stretch at the top is the one carrying 91% of the money.

That is where this stops being a capital expenditure story. A newbuild ordered this month does not arrive before 2029. A resale newbuilding already on the ways can be handed over inside twelve to eighteen months. HMM bought four hulls at a premium because it wants them working in 2027 and 2028, and the hulls it bought are the smallest in the package.

Here is the thing the coverage skipped, and it is the one that matters. Every write-up led with 104,000 TEU, because that is the big number and it makes the big headline. The money went the other way. HMM paid three to four times the per-slot price for 2,800-TEU feeders. Operators do not pay that for feeder tonnage unless feeder slots are what their own network is short of. The eight mainline ships are a 2029 decision. The four feeders are a 2027 decision, and they are the only part of this announcement that can touch a rate, a booking or a berth before the end of the decade.

The two halves are one decision, not two. A 13,000-TEU ship working a rotation is only as useful as the feeder that empties it, because every box it discharges has to be cleared by the berth, the yard and the network behind them. HMM is buying the ship that arrives and the ships that take the boxes away, and it paid a premium for the end of that chain that is scarce. If you read this as eight mainline newbuilds and stop there, you have read the press release and missed the purchase order.

The size tells you the rest. 13,000 TEU is not a head-haul ship. Asia-North Europe runs on 20,000 to 24,000 TEU. 13,000 TEU is the size that clears the expanded Panama locks, that works on Asia-US East Coast and on Asia-Mediterranean, and that can be cascaded down into a secondary string when the head-haul softens. HMM bought tonnage that can be moved between lanes. It did not buy tonnage that has to stay on the biggest lane to cover its own cost. That is a flexibility purchase, and it says HMM is not betting US$1.61bn on one trade still being strong in 2029.

That shape carries a consequence worth spelling out. Ships that size are not built for one service, because the lane they were ordered for is rarely the lane they earn on. The carriers that ordered into 2021 and took delivery into 2023 found out what a hull costs when the lane it was bought for is the one that collapsed. A 13,000-TEU ship that can move between US East Coast, the Mediterranean and a secondary Asia-Europe string is insurance against that, and part of the US$1.61bn is the premium on it.

The genuinely speculative part of this package is not the eight. It is the four. If HMM is right and 2027 feeder slots are tight, the resales were cheap at the price and the premium was money well spent. If HMM is wrong and the feeder market loosens, it owns four expensive small ships and a US$231m to US$366m lesson. That is a two-year market call placed three years out.

Put the two halves on a calendar and the shape changes again. Assume 30 months from signing to delivery on the eight: an order placed in the first half of 2027 lands in the second half of 2029, which sits right against the December 2029 ceiling and leaves no room for a slip. Assume 12 to 18 months on the four resales and those arrive somewhere in 2028. The package therefore has one half turning up in 2028 and one half turning up at the very back end of 2029, and the half that arrives in 2028 is the half carrying 8.9% of the TEU. Everything the headlines are about is the half that arrives last.

Layering it tells you who gets what. The eight big hulls will work the lanes a 13,000-TEU ship can physically enter: US East Coast via Panama, the Mediterranean, secondary Asia-Europe strings, and whatever they cascade into. If your port cannot take a 13,000-TEU ship, none of these eight will ever call at you directly, and the 104,000 TEU means nothing to you. The four feeders work the other way round. They will be on feeder legs and secondary ports, and that is exactly where a small addition bites, because a 2,800-TEU ship is a large share of a small lane. Ten thousand one hundred TEU spread across three feeder sizes is a rounding on a mainline lane and a real number on a feeder one.

Now the denominator. 114,100 TEU against roughly 1m TEU is 11.4%, and 11.4% is the figure that will get quoted for the next week. It is measured against a base that will not exist when the ships arrive. HMM's fleet in December 2029 will include whatever the 36 ships already on order have delivered by then, less whatever has been scrapped, less whatever chartered tonnage goes back. A percentage of a 2026 base is not a percentage of 2029 capacity. Treat 11.4% as an upper bound on a number nobody can compute yet.

There is a second denominator behind that figure, and it is the one a procurement team should be using. Eleven point four percent is measured against HMM's own fleet. It is not measured against the market you buy from. Spread 114,100 TEU across every lane HMM serves and every year from 2029 onward and it is a rounding on the capacity available to you. What makes US$1.61bn look like a market event is the size of the number, not the size of the capacity.

The sharper version of the same problem is the hull count. Alphaliner lists 36 ships on order. HMM announced 12. It did not say whether those twelve sit inside the thirty-six or on top of them. Inside, and this disclosure carries zero incremental capacity: it is a price tag on ships the market already knew about, and anyone repricing 2029 on the back of it is repricing on air. On top, and the orderbook reaches 48 hulls against 97 operating, which is 49.5%. Those two readings are three years and an entire rate cycle apart. Until HMM says which one it is, do not let anyone put this into a 2029 supply forecast with a decimal point in it.

There is a timing detail buried in the wording that reinforces the point. HMM said it will order the eight. It did not say it has ordered them. The December 2029 ceiling therefore depends on the yard slots still open on the day those contracts are actually signed, and it will move if the signing slips. A carrier that has not yet committed to a yard has not committed to a date either.

Why now, and why the money went where it went? Two conditions have to hold for a package shaped like this one. HMM has to believe 2027 and 2028 feeder capacity will be tight enough that waiting for a newbuild costs more than the premium, and it has to believe there are still 2029 delivery slots available at a price it accepts. The four resales are the visible cost of the first belief. The eight orders are the visible cost of the second. Neither one is a view on spot rates this quarter, and the announcement contains nothing that lets you read it that way.

For an importer or an exporter, the timing answer is brutally short. Nothing arrives in 2026. In 2027 and 2028 the only tonnage from this package that can exist is the 10,100 TEU of feeders, spread across three 2,800-TEU hulls and one 1,700-TEU hull, on whatever service HMM puts them. Against a 1m TEU fleet, that is 1%. If you move boxes through a Pearl River or Yangtze feeder leg, or through a secondary port that depends on a 2,800-TEU ship, that 1% is the only part of this story with your name on it.

So the eight hulls hand you a negotiating sentence and nothing else before 2029. The four feeders hand you a reason to go and look at your 2027 feeder and barge legs, because that is where 2,800-TEU tonnage lands, and that is where HMM just told the market it expects to be short. For anyone moving containers through a secondary Asian port, that signal is worth more than the 104,000 TEU everyone else is quoting.

There is a version of this where none of it reaches you, and it is the common one. If your lanes are worked by 8,000 to 14,000-TEU ships today, the eight newbuilds are the same size as what you already use and they add nothing you can see. If your feeder leg is worked by a barge network rather than by 2,800-TEU ships, the four resales will not be on it either. HMM runs a 1m TEU fleet across a great deal of water, and twelve hulls spread across it will be invisible on most of the trades its customers actually book.

The omitted lines matter more than the printed ones. HMM named no propulsion. A ship delivered in December 2029 will still be trading in the 2050s, and it will spend that whole working life under FuelEU Maritime intensity targets that step down on a fixed schedule, plus whatever the IMO finally lands on. A conventionally fuelled hull entering service in 2029 starts its career already behind the curve. That gap does not stay on the carrier's books. It reaches you through the fuel and emissions lines on your invoice, not through the base rate you negotiate. When 2029 carrier quotes come in, compare those surcharge lines separately. A low base rate attached to a dirty hull is not a low rate.

Let us do the arithmetic on a real book. Assume you ship 600 forty-foot boxes a year from East China to North Europe at a base ocean rate of US$2,000 a box. That is US$1.2m of base freight a year. Put HMM's 114,100 TEU against it. In 2027 the capacity this package puts on your lane is zero, so the rate effect is zero. In 2028 it is still zero, unless one of the four feeders lands on a leg you use, and then it is a rounding. In 2029 somewhere between none and all of the 114,100 TEU appears, spread across every lane HMM serves rather than yours.

So the money in this story is not capacity. It is posture. A carrier with twelve hulls to pay for and US$1.61bn going out the door by 2029 has a load factor it needs to defend, and a carrier with 48 hulls on order has one it needs badly. That is worth something at a contract table. On the book above, three points is US$36,000 a year and two points is US$24,000. That is the entire realistic value of this announcement to your 2027 budget: somewhere between US$24,000 and US$36,000, and only if you actually ask for it.

The case against reading it that way is real, and it is the same fact read from the other side. An orderbook of this shape is the bear argument as much as the bull one. If the twelve sit inside the thirty-six, nothing changed and the posture argument collapses along with it. If newbuilding prices fall before 2029, HMM overpaid for the four resales by the difference, and a carrier that has overpaid defends revenue harder, not softer. If those four feeders were bought against a specific service commitment, they will appear on that one service and nowhere else, which makes them worth nothing to anyone not on it. The condition that flips the whole reading is the thirty-six-versus-forty-eight answer. Everything else is second order.

Two traps to walk around. The 2029 ceiling is one of them: any 2027 or 2028 rate assumption that includes this tonnage is an assumption with a three-year hole in it, and the way you discover it is when the space you planned around never gets offered. Switching carriers on the strength of it is the other. A carrier with no orderbook is not automatically the safer 2027 bet. It is a carrier that has not yet decided what it will do in 2029, and when the market turns it may well be the one that blanks your sailing.

What you can do with it before the next quarterly disclosure costs one phone call. Ask your contract carrier in writing whether the twelve sit inside the thirty-six. If the answer is inside, every argument built on this capacity at the 2027 tender table can be refused on the spot. If the answer is on top, you have 48 hulls against 97 operating, and you can ask what load factor it intends to run to feed them.

A US$1.61bn orderbook is a confident document. It is also a three-year IOU, and IOUs get renegotiated. Price 2027 against the ships that exist today. If you want to lock something on the strength of this, lock the 2028 feeder legs, and do it before HMM tells us whether the twelve are inside the thirty-six, because that answer moves more than the money does.

↑ Back to the key points

— By Vivian Zhao