Containerization: the numbers
Pre-container baseline (1950s): Loading a mid-sized break-bulk ship took ~3 weeks in port with gangs of longshoremen hand-stacking sacks, crates, and barrels. Port labor was ~60–75% of total shipping cost. Ships spent more time in port than at sea.
The shift: Malcolm McLean's Ideal-X sailed in 1956. ISO standardization (1968–70) plus the Vietnam War logistics buildout (which forced the Pacific to containerize) drove adoption. By the 1980s break-bulk was effectively dead on major routes.
Cost impact estimates:
- Loading costs: Levinson cites a drop from ~$5.86/ton (break-bulk, 1956) to $0.16/ton (containerized) — roughly a 36× reduction. Bernhofen, El-Beialy & Kneller (2013) is the canonical econometric study.
- Port time: Ship turnaround fell from weeks to <24 hours, multiplying effective fleet capacity ~4–5×.
- Trade volumes: Bernhofen et al. find containerization raised bilateral trade among adopting countries by ~320% over 20 years — a larger effect than all post-WWII trade agreements combined (GATT rounds added maybe 45–285%).
- Total shipping cost share: Ocean freight fell from ~12% of US import value (1950s) to ~1–2% today. For many manufactured goods, shipping is now a rounding error.
Second-order effects:
- Pilferage collapsed — sealed boxes ended dockside theft, which had been a budgeted line item.
- Insurance rates fell sharply.
- Inventory costs fell — predictable transit enabled just-in-time manufacturing.
- Geography of production completely reorganized: it became cheaper to make a thing in Shenzhen and ship it to LA than to make it in LA. The China shock, the hollowing of US manufacturing, and the East Asian export-led growth model all sit on top of the container.
- Port geography shifted — New York, San Francisco, London (Docklands) collapsed as working ports; deep-water container ports (Newark, Oakland, Felixstowe, later Shenzhen, Singapore) replaced them.
- Ship size exploded — from ~1,000 TEU in the 1970s to 24,000+ TEU today, with per-unit costs falling further.
Why it shows up in TFP rather than just trade stats: containerization didn't just lower a price — it changed the production function of global manufacturing. Firms could split production stages across countries, specialize finely, and hold less inventory. Hummels (2007) estimates the implicit tariff equivalent of pre-container shipping was on the order of 10–20% for many goods; eliminating that is a permanent productivity gain.
The honest caveat: precisely decomposing containerization's TFP contribution from contemporaneous trade liberalization, jet freight, and telecoms is hard, and estimates vary. But it's one of the few late-20th-century innovations that everyone in the economic-history literature agrees moved the needle.