The $0.10 Freight Bill Inside Every Glass Jar
Direct answer
At August 2026 freight levels — 4,339perFEU∗∗onDrewry′sWorldContainerIndex,∗∗4,339 per FEU∗∗ on Drewry′s World Container Index,∗∗8,706 per FEU Shanghai–New York — a 500 ml glass jar carries roughly 0.10ofoceanfreight∗∗whentheemptycontainersareimported,againstroughly∗∗0.10 of ocean freight∗∗ when the empty containers are imported, against roughly∗∗0.0015 for a printed pouch shipped as film. That is a 65× gap on the packaging leg alone. If the product is filled at origin and shipped filled, the gap narrows to about 1.6× total shipped weight — still material, but not the headline number. Which of those two scenarios you are in decides whether switching format is a 5-figure or a 6-figure decision.
Why 2026 is the year this conversation came back
For most of the 2010s, freight was noise in a packaging decision — a few cents per unit, absorbed quietly. The packaging sector is now in what industry analysts describe as a structural cost reset, driven by sustained pressure on energy, raw materials and freight rather than a short spike.
The rate levels are not crisis-level, but they are not 2019 either. Drewry's World Container Index sat at 4,339perFEU∗∗inAugust2026,with∗∗Shanghai–NewYorkat4,339 per FEU∗∗ in August 2026, with∗∗ Shanghai–New York at∗∗8,706 per FEU, and SCFI Shanghai–North Europe running $3,500–4,500 per 40HQ in mid-2026. When a container costs four figures, the question "how much air and glass am I paying to move" stops being rhetorical.
Where freight attaches: the thing most comparisons get wrong
Flexible packaging and rigid packaging do not compete on the same shipping geometry, and most "glass vs plastic" articles quietly ignore this.
Rigid containers ship as containers. A 40HQ of cartoned 500 ml glass jars is volume-bound: the box structure and the jar's own shape fill the box long before you hit the payload limit.
Pouches ship as film. Printed pouches are converted from rolls at the packaging plant. You are not shipping the bag's empty volume; you are shipping a 100–120 micron sheet. That is why the arithmetic diverges by two orders of magnitude.
PET bottles already solved this — and that is the honest comparison. Good PET supply chains ship preforms, not blown bottles. A preform is a test tube of PET that weighs 20–30 g and stacks densely. So the right comparison is not "flexible beats all rigid"; it is "flexible and preform-based PET both beat glass, and flexible still beats preforms."
The container math, with assumptions on the table
Every number below is derived, not asserted. Change the assumptions and the conclusion moves — that is the point.
| Glass jar, 500 ml | PET bottle, 500 ml | Stand-up pouch, 500 ml | |
|---|---|---|---|
| Unit tare weight | 320 g jar + closure | 24 g (blown) / 22 g preform | 10 g laminate (100–120 µ) |
| Cargo form | Cartoned, volume-bound | Blown or preform | Printed film rolls |
| Units per 40HQ (empty packaging) | ~42,000 | ~1.1M preforms (weight-bound) | ~2,700,000 |
| Ocean freight per 1,000 units @ $4,339/FEU | ~$103 | ~$4 | ~$1.6 |
| Freight per unit | $0.103 | $0.004 | $0.0016 |
Assumptions: 40HQ usable volume 65 m³, payload 26 t; jar carton 400×300×155 mm, 12 units, 4.8 kg gross; pouch 160×260 mm with gusset at ~10 g; WCI $4,339/FEU (Drewry, Aug 2026). Your carton geometry will differ — the ratio will not.
Now the scenario that changes the answer.
Scenario A — you buy packaging, fill locally
The table above applies in full. You are importing empty containers, and glass is paying to move its own weight and air across an ocean. This is the situation for most beverage, sauce and condiment brands that co-pack in-market.
Scenario B — you fill at origin and ship filled
Now the product dominates, and the packaging is a share of gross weight:
| Filled glass jar | Filled pouch | |
|---|---|---|
| Product | 500 g | 500 g |
| Packaging | 320 g | 10 g |
| Gross shipped weight per unit | 820 g | 510 g |
| Relative freight | 1.61× | 1.00× |
1.6× is still the difference between a 20-tonne and a 32-tonne shipment for the same sellable units, and it compounds with every leg — ocean, port, DC transfer, last mile. But it is not 65×, and anyone quoting you 65× for a filled product is selling, not calculating.
Three costs that never show up in the rate quote
Pallet and warehouse cube. A pallet of glass jars holds fewer sellable units than a pallet of pouches, so your per-pallet storage and handling cost is higher at every node — 3PL, distributor, retail DC. For a 500 ml product, the difference is typically a factor of 2–3 in units per pallet position.
Breakage and claims. Glass in a multi-leg export chain needs heavier corrugate, dividers, and often shrink-wrap plus edge protection. Even then, a breakage rate that looks acceptable in a sample shipment becomes a real claims line at container scale — and a food-safety incident if a fragment reaches a filling line.
Disposal and take-back at destination. Retailers in several EU markets now charge for packaging volume in the back room. Empty glass goes out the same way it came in: heavy.
Where glass still wins, and should
We are a flexible packaging manufacturer, so treat the following as the part that costs us business — it is also true.
- Premium perception at the shelf. Spirits, high-end cosmetics, artisan preserves: the material is the message.
- Absolute barrier with zero migration risk. Glass is chemically inert. For aggressive solvents, high-acid products with long shelf life, or products where any flavour scalping is unacceptable, glass removes an entire category of risk.
- Refill and reuse systems with high cycle counts. A returnable glass bottle can run 20–30 trips. A pouch, even a well-built one, runs far fewer. Where a deposit-return infrastructure exists and the loop is tight, glass and rigid reuse win on life-cycle grounds.
- Hot fill without structure change. Glass tolerates fill temperatures that would distort many laminates.
The honest positioning is not "replace glass." It is: for single-trip, ambient, retail-shelf products where weight is a recurring cost, flexible is structurally cheaper — permanently, not opportunistically.
How to run this calculation for your own SKU
Five rows, ten minutes, no consultant:
- Tare weight per unit for each format, from the spec sheet (not from a brochure).
- Units per pallet, then pallets per 40HQ, using your actual carton dimensions.
- Ocean cost per container for your lane, current month — not last year's contract.
- Inland and DC handling per pallet position.
- Damage and claim rate per 10,000 units shipped.
Total those, divide by units. The format with the higher unit price often has the lower landed cost, and the gap is usually in rows 2 and 5 — the two rows that never appear in an RFQ response.
At NSTAR Packaging we quote against this structure rather than per-piece only: fill weight, carton configuration, and destination port go into the quote so the landed number is visible before you commit. Our food-grade doypacks run from a 100-piece minimum with 10–15 business day production after artwork approval, which means you can validate the freight math with a real shipment instead of a spreadsheet.
FAQ
Is flexible packaging always cheaper to ship than glass? For empty containers, yes, by a wide margin. For product filled at origin, the advantage narrows to the packaging's share of gross weight — typically 1.5–1.7× for a 500 ml product.
What about PET bottles — aren't they already light? Blown PET is light, but the structural win is shipping preforms and blowing in-market. Brands that ship blown bottles are leaving most of the advantage on the table.
How much does a 40HQ actually hold? Practically 65–68 m³ usable and about 26 t payload. Which limit binds first — volume or weight — is exactly what determines whether your format is efficient.
Does a lighter pack really reduce damage? It reduces handling damage from weight and drops, and eliminates shatter. It does not remove puncture risk, which is why laminate selection and gusset design matter more than thickness alone.
Can a pouch survive a hot-fill product? Retort and hot-fill capable laminates exist (typically PA or foil-bearing structures). They cost more and change the recyclability profile, so specify them deliberately rather than by default.
What freight rate should I model? Use the current month's index for your lane and add 15% for surcharges and inland. In August 2026 that meant 4,339/FEUonWCIand4,339/FEU on WCI and∗∗8,706/FEU Shanghai–New York.
Sources
- Drewry World Container Index, August 2026 levels, as reported in CargoDuo, Ocean Freight Rates 2026: Where the Market Stands, 21 August 2026 — cargoduo.com
- SCFI Shanghai–North Europe levels, July 2026 — greathensen.com
- Packaging Gateway, Packaging sector faces structural cost reset, 13 April 2026 — packaging-gateway.com
- Toward Packaging, Glass-to-Plastic Packaging Market — lightweighting and shipping-cost dynamics — towardspackaging.com
Want the landed-cost comparison run on your actual SKU? Send your fill weight, current jar or bottle spec, carton dimensions and destination port. We will return a side-by-side per-1,000-unit landed cost with the freight assumptions written out.
