Six years ago, early in my tenure at Berkman, I wrote a blog post that tried to calculate the cost of shipping water from a bottling plant in Yaqara, Fiji to Cambridge, Massachusetts. I was interested in unpacking the everyday mystery of container shipping – how is it possible that we can sell a product for a couple of dollars a bottle despite shipping it 8,000 miles around the world – and in the odd idea that atoms might be more mobile than bits, as we get lots more Fiji water in the US than Fijian music, movies or news.
My estimate then was that a 40′ container filled with Fiji water would cost roughly $5000 to deliver from Suva, Fiji to Cambridge – I came up with the estimate based on a variety of statistics about international shipping that I bent and welded into a Fiji/Massachusetts estimate. At $5000 a container and 24,000 kilograms per 40′ box, it would cost $0.21 for a liter bottle of Fiji water to make the 8,000 mile journey. Not free, but a small fraction of the retail price of a bottle of “premium” imported bottled water.
I had occasion to return to this blogpost today – I’m working on a book, and this Fiji example features in it. So I decided to recalcuate the numbers and see if I could find an answer that’s more defensible and satisfying.
Turns out I got a few details wrong. First, the 24,000kg figure applies to smaller, 20′ containers – the limit for 40-footers is 30,480kg. And the price from Suva to Cambridge for a 40′ container is just slightly higher – $5,540.30. That comes out to $0.18 per liter, three cents less than I calculated six years ago.
These new figures come from my new favorite toy, Maersk’s online shipping rates calculator. The Danish superfirm A.P. Møller – Mærsk Gruppen is the largest shipping group in the world, with offices in 135 countries, 120,000 employees, and roughly 600 container ships, capable of carrying more than 2 million 20′ containers at any given time. They’ve also got a thoroughly badass IT system, which they’ve now made accessible to the general public.
Okay, it’s not exactly Amazon.com, or even Fedex. To use Maersk’s calculator, you need to register with the site, download a client browser certificate and accept three server certificates from Maersk before you can access their secure site. But once you do, it’s just a few short clicks before you can calculate the cost of shipping a 20′ container of “umbrellas, sun umbrellas, walking-sticks, seat-sticks, whips, riding-crops and parts thereof” (yes, that’s one of the available categories, along with “bone and meal”, “ores, slag and ash” and “straw, esparto, other plaiting materials and articles of straw, esparto, other plaiting materials) from Auckland to Dubai: $2451.02
The main thing I’ve found playing with Maersk’s calendar: distance doesn’t matter as much as demand. Americans buy a lot of atoms from China. The Chinese don’t buy nearly as many from the US. A 40′ container filled with household goods, shipped from Shanghai to Houston, TX costs $6169.93. Reverse the trip and ship the same container from Houston to Shanghai and the cost is $3631.07. That’s because 60% of containers on ships coming from the US to China are empty, which means Maersk and other shippers are desperate to sell container space.
(The 2006 New York Times article that offers that 60% empty container statistic suggests that lots of full containers are coming to China from raw-materials rich countries like Australia, Brazil and the Middle East. That suggests we should see the opposite pattern – expensive containers from Sao Paolo to Shanghai and cheap ones in the other direction. Nope. $5101.70 from Shanghai to Sao Paolo, $1930.59 in the other direction. Perhaps containers from China to Brazil are riding the same ships as those to the US and paying the same premiums?)
Maersk also offers a set of maps that help you get a sense for how these trade routes actually work. It’s a four day trip from Suva to Auckland on the Pacific Islands Express, and then the bottles of Fiji water are transfered to OC1, the Oceania Americas Service. The Pacific crossing is a long one – 18 days to the Panama Canal, a quick stop in Cartagena, and we’re in Philadephia 25 days out of Auckland. It’s a truck ride from Philly to Cambridge, and that short hop is responsible for $950 of the total transit cost.
As I poke through these maps, schedules and tariffs, I feel like I’m glimpsing a secret world. Part of it may come from the sheer poetry of the names. Shipping routes include “The Boomerang” and the “The South China/Australia Yo-yo” and connect ports like Tin Can Island (Apapa, Nigeria, the main port for Lagos). And part comes from the sense that these routes and rates, the infrastructure that supports an economy where transPacific bottled water is possible, are the ley lines of globalization, radiating a mysterious and sinister power.
Well, on that analogy with the Sao Paolo – Shanghai route: Raw materials aren’t shipped in containers (and therefore not on container ships). It’s a different market.
adiannamalai, a village 6 km away from tiruvannamalai, piles of tomatoes rotting, because the cost of “shipping” them to the market by bullock cart is higher than the price they can be sold for
my point is, transportation facts/figures are infinitely interesting
Joerg, I’m not sure that’s true – Maersk’s system seems to indicate that ores and other raw materials are frequently shipped in 40′ boxes. If I understand Brazil/China trade, soybeans and beef are often exported from Brazil to China, which are also containerized cargos. I’m sure you’re right that oil is a different market, but I do think there are some interesting open questions around the container market on that route.
Gregory, I’d love to offer some sort of map of comparative travel costs. The Maersk data makes clear to me that container shipping is lots less expensive than trucking, mile per mile. My guess is that rail is somewhere in the middle – CSX offers a pricing calculator at http://csx.com/index.cfm/customers/prices-tariffs-fuel-surcharge/ that appears to be down at the moment, but I hope to compare trucking, shipping and rail sometime soon. But yes, the more rural we get, the less infrastructure we have, and the more likelihood we end up with rotted tomatoes and poor farmers… I find myself wondering whether refocusing development aid on building infrastructure in the developing world on a massive scale would have more impact that the work that’s been done in the past.
Just curious if you’re taking into account the cost of clearing foreign water through customs? Someone like Maersk would typically not be responsible for that type of thing, so I doubt they include that in their calc. It may be a neglible amount on a cost/bottle basis, but still a critical step.
Ethan: As far as comparing shipping mile-per-mile to trucking or rail, be aware that ship costs for a transit between a foreign port and a domestic port are much cheaper than ships costs for a transit between two domestic ports. Jones Act requires all vessels traveling between two US ports to be US flagged and US crew. Maersk does own and operate these types of vessels, but their rates will be much more expensive than those for importing/exporting.
Thanks, Dan. Maersk’s data includes a large number of tariffs – including a submission of cargo declaration data fee and a documentation fee – which may address the customs charges. If not, I’d be interested in learning more to get a sense for how they’d affect the price per bottle. Thanks for the insight on the Jones Act – actually, Maersk’s tool allows you to select the flag of the vessel, which might allow for some of these calculations to take place domestically.
Interesting post. I often get into this topic with locavore neighbors who like to think in terms of food miles. Sometimes I have hazarded a guess that, for example, shipping a leg of lamb from New Zealand to Washington State uses up very little of the world’s resources, despite all the food miles. Extrapolating from your numbers, it looks like I might be right that if the NZ lamb arrives in Seattle by container (100 road miles from us) total transport costs could be less than lamb delivered 300 miles from a rancher in Eastern Washington. Now I need to find out how much it costs to ship a bunch of asparagus to Washington from Peru. Then I’ll be ready for my next dinner party.
Pingback: Distance isn’t dead [The Economist] | DreamInn
Ed – I wrote a piece a couple years ago for Worldchanging that pointed specifically to the New Zealand lamb example, and suggested that some aspects of the food mile concept might be interpreted as a form of nativist protectionism: http://www.worldchanging.com/archives/006189.html But for actual answers on this question, I’d recommend Jason Clay, who thinks through these sorts of issues for the World Wildlife Fund. Notes on a fairly recent lecture from him here: http://ethanz.wpengine.com/2009/07/04/jason-clay-and-measuring-the-environmental-impact-of-agriculture/
Pingback: The Ley Lines of Globalization: Analyzing Shipping Freight Around the World | The Global Warming Statistics
I wonder if the 60% figure is still accurate (since it’s from a 2006 article), and how it fluctuated over the recession.
Great post – I find this extremely interesting. And mirroring the sense that you are getting insights into a “secret world”, I remember volunteering for an NGO in Northern Mexico, going with some of the NGO members to the port to pick up a shipment of crutches and wheelchairs from the Netherlands – while waiting for the port official, I picked up some of the trade magazines lying around, which were all about trade routes, scheduled departures, new equipment at ports, etc. Can’t wait for your book!
Container rates are related to volume and weight or how much a container can hold. And, historically, rates are also related to the market price of the good. Shipping costs have to supportive of low value products needed to fill ships and support trade. And, we can’t forget the military/industrial complex.
The trade volume effects rates through competition and economies of scale. Trade imbalance is reflected in inbound and outbound rate differentials. It is better to reposition a container with at least some revenue against fixed costs than nothing. The more and more complex vessel/port networks route also have an effect related to trade imbalances.
To understand container pricing it’s all about fixed and variable costs verses vessel utilization and revenue. Shipping is a very high fixed cost business.
Shipping will always be much cheaper the rail or road as it doesn’t have the higher costs of their networks as well as the energy and machinery required to move a ton of freight. Water highways only have canal and port costs.
This is really cool, Ethan. Since reading The Outlaw Sea by William Langewiesche a year or so ago I’ve been fascinated with the idea of how much work and sacrifice goes into getting, say, an iPod from a factory in Dongguan to a Best Buy in Albuquerque. It’s staggering, and really one of the untold stories of our age.
Pingback: Don’t Bet on China | The Libertarian Standard
Ethan, one of the nuances to importing water specifically is that you would have to prove that you aren’t carrying any foreign contaminants. There would probably be some additional steps on top of the tarifs you pay for the cost of the goods. If a bottle of Fiji water is chemically treated, then you may be fine, but you would still need to prove that to Customs. I have know idea how rigorous USCBP is with this stuff, but I wouldn’t be shocked if they require random testing, etc. This seems simple in theory, but in reality, this means you’d probably have to hire someone to help you out with this.
Just to be clear, Dan, I’m not trying to figure out the entirety of Fiji water’s cost structure – I agree that there are customs and inspection costs involved. My goal was just to try to get a general sense for what chunk of Fiji’s costs come from transport.
Pingback: The week that was | pendulously
Pingback: Trinkets « Trinkets Of Frivolous Utility (TOFU)
I have often wondered about this in relation to internal trade in Brazil. The roads are very poor, dangerous and need billions spending on repairing/modernisation.
There seem to be endless trucks coming from both north and south to the city of Vitoria. The city has a container port which, as far as I can see, only deals with imports and exports.
It would make a lot of sense for some of the food and manufactured goods to be transported by sea.
While coal, iron ore, grains and other dry bulk cargoes can be carried in containers, they generally aren’t – and not in the quantities or along the distances you are thinking of. Dry bulk cargoes are carried in ships called (smartly enough) “bulk carriers – and they range in size from very small single hold coastal carriers to massive, 7-9 hold, 300,000-400,000 deadweight ton VLOCs (Very Large Ore Carriers, also very smartly named). These ships carry everything from grains to wood chips (for paper etc) to bulk minerals. Smaller cargoes, or cargoes bound for smaller ports that cant take larger carriers are generally shipped as break bulk on “twin decker” cargo ships, and so also aren’t likely to end up on container ships. Equally, because of the corrosive nature of many bulk cargoes, the containers generally have to be lined (or the cargo bagged) before shipping, hence further increasing workload – and inefficiencies of container shipping vs bulk shipping.
While containerization has greatly increased the efficiency of shipping – it is really only suited for shipping finished and packaged products. Raw and processed materials are almost always shipped as bulk (non-containerized) cargo – whether it be coal, crude or refined oils, orange juice concentrate, or otherwise. Indeed, even many finished products such as cars, which can be shipped in containers, are cheaper to ship on specialized ships (in this case Pure Car Carriers – PCCs). Sorry for the length, I’m in the business.
Pingback: The Ley Lines of Globalization
Ethan – post much enjoyed – highlighted at Bowwave.com
A shipping e-zine
Like Young James, I’m in shipping.
I recall a couple of seasons – perhaps eight or ten years ago (I’m guessing, but ‘turn-of-the-century – ish) when some bulk commodities – coal, iron ore, maybe bauxite – were shipped in open-top containers on some runs because:
– capacity was there
– loading/unloading was reasonably cheap [either well mechanised or cheap labour}.
I don’t think this lasted very long – seasons, not years.
I do recall, perhaps twelve years ago, reading that the cost of shipping a bottle of Scotch [from Scotland – obviously] to Hong Kong, was 6 [British] pence [say US 10 cents, then as now]; less than the cost of the label on the bottle!
I think the proviso was that five Teu were shipped every month, but as boxed Scotch takes up much more space than bulk Scotch, and as Hong Kong seems – at least, when I was there – to have a reasonable market for the stuff – that may have been easy to ensure!
And the speific fuekl usage of a large ship -even a twenty-five knot containership – note – per otnne-mile – carried is hugely low compared with trains or trucks or planes.
I do wonder if there is space in mch of the world for a better developed canal system [Continental Europe has this; is it feasible in east Asia or much of Africa?]; once the goods are in transit, a barge arrives every day . . . ]
Re tgalarneau’s comment –
” Shipping is a very high fixed cost business.”
– Shipping is cheap to enter in a small way; if successful, finance has, historically been available to expand.
Is this stil true?
Shipping can be an asset play on hulls [which may earn freight in the time between buying and selling/selling for scrap].
Oh – the assets can be amortised over twenty years for some ships – although not serially abused bulk carriers.
There are many ways to turn a buck in shipping – I have hardly touched on them here – and I am sure there are many I don’t know, probably more than I do! – despite having been in merchant shipping since it was legal (if not, really, ‘the done thing’) to hang malefactors at the Yard arm in the – British – Merchant Navy].
Pingback: DYSPEPSIA GENERATION » Blog Archive » The ley lines of globalization
Pingback: What we learned this week « Make Wealth History
The modern world is growing at a rapid pace. 30 years ago, no one even knew what the Internet was, and few people had access to a computer. In todays world the internet technology has revolutionised the way that we spend our time, from internet TV to internet shopping and on-line games. A lot of the shopping we do can be completed online and paid for from the comfort of our own home, and the choice of products and suppliers is endless, we can now place an order with a supplier on the internet, pay for it and have it delivered the next day, Next day parcel delivery from internet shopping has now become a way of life, and is taking over from the traditional high street, with no parking problems and no walking around in the rain,
With the internet has also come the demise of the “Yellow Pages” no longer do we look in the “Yellow Pages” for services or products, we just go to a search engine like “Google” enter the product or service we require and we get a list of may be 1,000,000 suppliers from all over the world, with each supplier providing us with prices, descriptions and images of the products and giving us a much wider range of choice than we would have had 30 years ago.