Half the world’s cargo ships sail empty—could there be an app for that?
The last time I read about ballast was as a teenager learning Canadian history. Ships carrying furs or timber to Britain needed ballast coming back to Canada and ironware, rocks and people served that purpose nicely. The real money was made on the eastward voyage, but to go east you first had to go west, and if carrying people helped you sail west more smoothly and safely, why not take them? Plus, people, though not as heavy as rocks, were willing to pay. Hence Canada got settled in part because ships heading west for raw materials needed ballast.
Ballast hasn’t come up again until this week. (I happily concede that’s wholly a result of my own parochialism—some people presumably make their careers out of ballast.) What brought it to mind again is a new research paper by my McGill colleague, Theo Papageorgiou, and two other economists (Giulia Brancaccio of Princeton and Myrto Kalouptsidi of Harvard). Their paper, just published by the National Bureau of Economic Research, is called Geography, Search Frictions and Endogenous Trade Costs.
People modelling international trade typically assume transportation costs vary according to the distance travelled, which seems reasonable; fuel and wages are important costs and both rise as the trip gets longer. But Theo and his fellow researchers figured it was probably more complicated than that and decided to try to model the market for dry bulk carriers, that is, ships of various size into which cargo such as grain, iron ore, coal and alumina ore is poured directly and hauled away.
It’s a very interesting market. Some 10,000 ships roaming the seas like taxicabs, hooked up with exporters by brokers, with no dominant carriers. The average contract is written six days before loading. The average trip lasts 2.5 weeks. The average fee is US$10,000 a day and $140,000 per contract (though that doesn’t quite figure, does it?). What makes the study possible is satellite data on ship movements. What a century we’re living in! GPS can tell the ship’s draft, i.e. how high or low it’s riding in the water and therefore whether it’s full or not.
It turns out lots of ships are sailing around empty. About 45 per cent of them at any one time. The reason is not that the market is inefficient—though there may be some of that, as the researchers try to investigate—but rather that it’s the nature of what is often one-way trade. Lots of the full ships are headed to China, carrying raw materials that are going to be transformed into manufactures. As everyone in the world knows, thanks to reminding by President Trump, China exports tons of manufactures. But it does so in container ships, not dry bulk carriers. (You don’t want to just throw all those computers and flat-screen TVs into the bottom of a hold.)
As a result, dry bulk carriers make money going to China (and similar inhalers of raw materials) but often leave empty. What they do next is shop around (or ship around) for cargoes—goods they can pick up nearby and take to wherever they’re ultimately heading to pick-up another cargo to take back to China on the more profitable leg of the trip.
Theo and his colleagues provide lots more interesting information about the market and build a “search model” of the sort economists often apply to labour and taxi markets, which they then “estimate” using the, ahem, boatloads of data they have about trips, contracts, ships and so on. (“Estimating” means they try to see whether the model explains the different patterns they see in the data.) I’m afraid I acquired my econometrics closer to the Age of Sail than they did so have some trouble covering all their numerical tacks. What I do get out of it is that, like so many markets, this one is complicated.
Changes on the main dry bulk routes—whether because of changes in fuel costs or slowdowns in the Chinese economy—will have knock-on effects on all these ballast routes and countries and elsewhere besides. Fewer ships going to China means fewer ships stopping in ballast countries, too. So there are repercussions all round.
Another simulation they perform is to see what happens if shipping costs between Northeast America and China fall markedly (30 per cent, to be precise), as they would do if the Northwest Passage opened up to general navigation. Brazil and Northwest America are hurt because ships now prefer to ballast to Northeast America. Interestingly, bulk exports from China fall slightly—ships that have brought goods there now have more options about where to go next and don’t have to cut their prices as much, thus discouraging Chinese exports.
Is having so many empty ships sailing the high seas efficient? It can be—there’s only so much money you want to spend looking for ballast customers. But the researchers’ econometrics suggest it isn’t entirely. In 2017, that naturally makes you think that maybe there’s an app that will help get the unused capacity used, a kind of Uber for the oceans. It will be interesting to see what the numbers say in another decade.