Pandora’s Shipping Container

The world economy has grown by about 19x since 1950. Total trade has grown by around 60x. This means the world economy has gotten much more globalized since the middle of the last century.The Box is the story of how. The short version is that shipping used to be

The world economy has grown by about 19x since 1950. Total trade has grown by around 60x. This means the world economy has gotten much more globalized since the middle of the last century.The Box is the story of how.

The short version is that shipping used to be inefficient because ships got packed by hand. A team of longshoremen would empty a ship and fill it up with discrete goods, deciding ad hoc what would go where. This required a lot of time, manual labor, and even computation — packing a ship full of heterogeneous objects is an NP-hard problem. By contrast, packing a ship full of uniform shipping containers is a pretty easy problem. Like playing a version of Tetris where the only shapes are 2x2 blocks.

That’s exactly what happened starting in the late 1950s, accelerating throughout the next half-century.t. Small shipping lines retrofitted ships so they could hold nothing but boxes. Ports got redesigned to accept incoming boxes, quickly remove them, and drop them directly onto trucks and trains. Manufacturers learned that the cheapest way to receive materials or ship finished products was to do so in 8’x8’x20’ or 8’x8’x40’ increments, so they rescaled their businesses accordingly.

There were some winners: China (by a mile), Singapore, Long Beach, Newark, Rotterdam. There were losers: New York, London, Liverpool. Whether or not the entrepreneurs won depends on when you take your snapshots. The shipping business has always been cyclical, and as the cost of loading and unloading compressed, while capital costs rose, the business got more cyclical as more costs got fixed.

These stories are always easier to tell if there’s an actual story, so roughly 50% of The Box is about the general rise of containerization and 50% is about the career of Malcom McLean of Sea-Land, a shipping pioneer who made a fortune, nearly lost it, made another, sold out, bought back in, and eventually lost almost everything.

“Conflict among executives as a given; managers were expected to meet, thrash out their differences, and act. Performance was measured constantly, and rewarded not with cash but with stock in the fast-growing company.” McLean was building a business with low incremental costs coupled with high capital costs and strong network effects; remove enough specific language and it sounds like a software company.


A recurring theme in the book is the importance of labor and labor unions. Even before containerization, McLean makes his initial fortune in trucking in the South. At the time, Southern workers were much less unionized than in the North, which led to an absolute cost advantage that grew during economic downturns.

In the 50s and 60s, longshoremen were strongly opposed to containerization, because it reduced the demand for their labor. In an odd turn of events, some longshoremen’s unions actually negotiated a Coasian bargain: in exchange for letting ports containerize and loosen work rules,which would gradually cost the longshoremen their jobs, they accepted lump-sum payments for their forgone income.

(A wonderful bit of historical irony: immediately after the bargain, longshoremen’s unions started lobbying hard to automate the jobs faster. The shipping companies responded to their new incentives by giving the longshoremen more, and heavier, loads to carry. The faster the longshoremen could replace backbreaking labor with cranes, the happier they were.)

This was not uncontroversial. Everyone from the mafia to Eric Hoffer registered qualms about the deal. It eventually passed, but only because the players had different time horizons. The unions thought through their members’ entire careers; the companies thought through eternity.

This sounds like an exaggeration, but it’s not. A corporation, like a naked mole-rat, is effectively immortal. It can die, but its mortality rate is constant over time.

Both unions and corporations pool risk and reward, and coordinate efforts. The big difference between them, traditionally, is that one represents labor and another represents capital. But another important difference is that a share of a company can be sold and transferred, while union membership largely can’t. If a company does something that leads to a high present dividend at the cost of a larger value erosion in the future, shareholders suffer. If a union does something that raises current members’ incomes but causes the union to shrink as its members retire, there’s no constituency to argue against it (except Hoffer, who argued from a moral standpoint, and the mafiosi, who do think of the union as an immortal asset they’ll transfer to their kids).

This pattern will likely persist over longer cycles of economic history: when property rights are certain, we should expect more assets to end up in the hands of effectively immortal institutions: colleges, religions, families that practice primogeniture, and of course corporations.

Divergence Over Convergence (Excluding East Asia)

Simple economic theories argue that if you compare two countries with similar resource and population endowments, the poorer one should grow faster. A poorer country has cheaper labor (so they can compete on price) and can skip wasteful expenses (going from no phones to cell phones with no land lines in between), which in turn raise investment returns, allowing the poor country to import capital from the rich one.

This is not what happened in the twentieth century: by and large, rich countries got richer relative to poor countries. Standards of living in the US and Western Europe have improved relative to Africa over the last few decades, even though Africa has more low-hanging fruit.

The Box explains some of this, albeit indirectly.The reason Africa and South America didn’t catch up is partly because a few Asian countries did.

There are three basic reasons for this:

  1. These countries already had ports, and decent transportation infrastructure to get goods to ports, so they could take advantage of cheaper shipping. (Specifically: Japan had railroads; China placed one of their Special Economic Zones near a good port; and in Korea, Park Chung-Hee happened to get a lot of votes from a region that was also near a port, so he put one of the world’s largest steel mills there.)
  2. They made better policy decisions.
  3. Dumb luck: they were next to Vietnam.

Vietnam as a driver of globalization is one of those ridiculous contingencies that happens to completely change the course of history. The US deployed lots of troops to Vietnam, but couldn’t adequately supply them. Rather than do something rash, such as not getting more involved in a land war in Asia, the Pentagon gave a major container shipping company a contract.

This solved the supply problem: Sea-Land shipped vehicles, uniforms, guns, ammunition, and MREs to Vietnam. Vietnam didn’t have a lot to ship back, but the contract was plenty lucrative even if the containers were completely empty on the return trip. However, they didn’t have to be: Japan was in the middle of an export-driven economic boom, and they were quite interested in using containers. So Sea-Land’s West Coast to Vietnam trips started making stops in Japan. Essentially, the logistics of the Vietnam war were an indirect subsidy for Japanese manufacturers, which they took full advantage of.

(To give credit where credit is due — and when you’re talking about Japan, credit is generally due to MITI — Japan’s industrial policy bureaucrats had already decided that containerization was the future.)

During the Vietnam war, Japanese exports were a nice bonus. After the war, container shipping from Japan still made sense, so the ports of Seattle, Los Angeles, and Long Beach continued to shift to containers. And when Korean exports stepped up in the 70s and 80s, and Chinese exports stepped up in the 90s and (to an unprecedented degree) in the 00’s, we were ready.

World history could have been very different if we’d decided to fight an endless, unwinnable war in the Congo instead.

Cheap Shipping is its own Trade Policy

Basic economics tells us that if we raise taxes on international trade, we get less of it. Economics also tells us that if we relabel a “tax” as something else, it doesn’t affect behavior.

So, when the cost of shipping goes from ~20% of the value of the goods to ~0% of the value of the goods, that’s equivalent to a massive change in trade policy. This is a helpful way to reframe the current trade debate. From one perspective, we’re imposing unusually high tariffs on China.[1] . In reality, high tariffs just return us to the pre-containerization status quo.

That may or may not be a good decision. Perhaps in 1960, the optimal tariff policy was a 95% rebate on all shipping costs. Perhaps not, though.

Globalization can act as a substitute for technology as a source of growth. If we can’t raise American standards of living by producing new products, we can theoretically raise them by selling more of our stuff to developing nations. But technology is a complement to globalization as well. A new technology — better shipping, in this case — allowed the world to produce much, much more. But what we produced was more of the same.

For poor countries (at least the ones that build container ports and factories), that’s a boon. But to rich countries, it’s not. We all compete with China for the same oil, iron, and copper, and the outputs from the same finite supply of arable land. Uneven technological development can actually reduce rich countries’ standards of living. If the Internet, global logistics, and finance have all made it easier for other countries to bid for the same raw inputs, the price of those inputs will rise. And given that there are 330m Americans and 5.8bn people in the developing world, the “tax” from higher demand may well outweigh the benefits from higher efficiency.

The globalization tax compounds. Every year, the stock of available natural resources declines, while cumulative carbon emissions increase. When resource stocks decline, resource costs become a larger share of total costs relative to labor and capital. Higher depletion reduces the elasticity of supply with respect to price, too: when resources are abundant, it’s pretty easy to step production up or down. But now, opening a new copper mine is a serious engineering project.

The status quo is gradual decline.

The solution to this is a more even pattern of technological growth. If we’re able to consistently generate more and better goods from a given set of raw inputs, technology wins; if we can only increase productivity in a limited subset of high-margin mostly-virtual goods like software and financial derivatives, the benefits of that innovation are gradually diluted by globalization.

Holding everything constant and thinking in a purely utilitarian sense, this is fine.Globalization means Americans spend more on gas because more people in Indonesia get their first car. But in the long term it’s destabilizing. It means control of resources becomes more important over time.

We’re already approaching that point. While the Belt and Road Initiative is not going so great, the possibility that China will cut off rare earths is a sign that real resources constraints matter. We should think seriously about what China’s incentives are, and what the US’s options are, if globalization continues to outpace technology as a source of worldwide economic growth.

The world has prospered mightily from cheaper shipping. It gave us decades of faster economic growth, fueled by developing countries in East Asia catching up to the developed world’s growth. It benefited American companies, by allowing them to outsource labor-intensive manufacturing while continuing to service their home markets. It allowed them to take advantage of American R&D capabilities and capital markets.

But that’s ultimately not the kind of growth the economy needs. If The Box were one of a hundred such books, each about a 20x improvement in technology — cars getting 500 mpg, planes flying at Mach 8, cheap palladium from asteroid mines, higher farm yields via geoengineered climate change, some sort of hypothetical fantasy power source that was safer and more energy-dense than fossil fuels — The Box would be an undeniably happy story.

It’s not, though. It’s a story about a technology that allowed other countries to copy the US and Western European manufacturing model, while giving the developed world increasing prosperity without the need for productivity growth.

Higher tariffs are a return to the pre-container status quo. The US was hardly a poor country when we couldn’t outsource our way to incremental prosperity. We were a country that would get richer if it invented new and better products.

Ctrl-C/Ctrl-V prosperity is reaching is natural limit, as the Workshop of the World transitions into the Armory of the CCP, and nation-states get more aggressive about securing scarce resources in a zero-sum race.

It’s time for Ctrl-N.