A container that sits still at a port or a transshipment hub costs a smaller exporter the delivery window and the extra inventory held while it waits. Port and logistics infrastructure matters because the public clocks time the ship, the container and the scheduled network separately, and the charge for one shipment stays on that shipment’s own papers.
Port and logistics infrastructure matters to a smaller exporter as ship time in port, container dwell and the scheduled liner links that reach the loading port. The World Bank and S&P Global methodology note of 10 June 2026 times a container ship from arrival at anchorage, the pilot station, or the port limit until the ship leaves the berth, and it excludes the time after the ship leaves the berth. Connecting to Compete 2025, cited as a 2026 World Bank publication, reports 21 logistics indicators from shipment tracking for 2023 and 2024, and states that cost is not directly measured. The WTO World Trade Report 2015 says fixed costs of entering a foreign market impinge particularly on the profits of small firms. Days and charges for one booking stay blank until that booking’s documents name them. Related explainers sit in the World section.
| Item | What the saved page prints | Why it matters |
|---|---|---|
| Ship time in port | CPPI note, 10 June 2026: from arrival at anchorage, the pilot station, or the port limit until the ship leaves the berth. Time after the ship leaves the berth is excluded. | This is the ship’s clock, scored for the port as a whole. |
| Container export dwell | LPI 2.0: from arrival of the full export container until loading on the ship is complete. Buffer time is included. The printed example is 48 hours before departure. | A cut-off set by the line can sit inside this interval. |
| Container import dwell | LPI 2.0: from unloading until the container leaves the port. | The report ties this interval to clearance, terminal work, trucks and infrastructure. |
| Liner connectivity | UNCTAD LSCI, country or port. The metadata file lists six scheduled-service components and prints two bases for 100. | A scheduled link is the network on the timetable. |
| Cost | LPI 2.0: cost is not directly measured. | The freight invoice is a separate paper. |
| This route’s days and price | The saved pages do not print them. |
Where an export shipment spends its time
The container and the ship keep separate records. Connecting to Compete 2025, the World Bank report that introduces the Logistics Performance Indicators 2.0 (LPI 2.0), defines the container stages in days and says the indicators are reported by economy.
Export supply chain initiation runs from an empty container being sent to the exporter until the shipping line takes the full export container in charge at the port or an inland facility. The report says the two indicators built from empty-container movements reflect how shippers, consignees and trucking companies work together.
Container export dwell time runs from the full export container’s arrival at the port of departure until loading on the ship is complete. The report says this interval includes buffer time, voluntary or mandated by the shipping lines, and gives 48 hours before departure as the example. It also says formalities such as certificates of origin, processed upstream, are not automated in many economies and may delay containers.
The ocean leg comes next. LPI 2.0 counts the number of transshipments and the time spent in them. The report says about half of containers travel directly to their destination, and that the rest may be transshipped, up to eight times in some cases. It says each transshipment adds to maritime lead time, roughly 18 extra days per transfer (figure 2.3, panel b). On one printed route, Hamburg to Santos in 2024, the report counts 3,806 completed voyages, of which 168 containers were transshipped once, in Morocco, the Netherlands or Spain. The prose around that figure does not print the median days for each phase, so those day cells stay blank here.
At the destination, container import dwell time runs from ship unloading until the container leaves the port. The report says import dwell and delivery times capture customs clearance and border management, seaport and airport operations, cargo removal, the availability of trucks or other equipment, and the quality of transport and logistics infrastructure. Import procedures are generally more complex and time-consuming than export procedures, except in economies with strict export controls.
For a landlocked developing country the report defines a comprehensive container import time as the sum of three indicators:
LLDC import time = dwell at the coastal port of arrival + corridor lead time from port exit to the destination + dwell at the destination clearance facility.
That sum is the report’s definition. It does not print one country’s three day counts in the prose used here, so a numeric example stays blank. The report does say corridor lead time and dwell at the coastal port of arrival tend to be the largest of the three, followed by dwell at the destination.
The ship’s own clock is the Container Port Performance Index (CPPI). It starts at the ship’s arrival and stops when the ship leaves the berth. It is drawn under the container’s path because both can be running while the ship is alongside, and they answer different questions.
Takeaway: Name the stage before you call a route slow. The report’s stage definitions are the map. The days for your booking are not on that map.
Port efficiency, connectivity and dwell time are different clocks
The 2025 CPPI publication page says the index measures the total time a container ship spends in port, including time waiting and time at berth. The methodology note of 10 June 2026 is more specific. Port time includes all time from a vessel’s AIS-detected arrival at the port limit or anchorage zone until the ship leaves the berth. Key event 1 is arrival at anchorage or the pilot station. Key event 5 is departure from the berth. The time between event 5 and event 6, after the ship leaves the berth until it exits port limits, is excluded. The note says operations after the ship leaves the berth, such as bunkering, repairs, or waiting in a safe area, are excluded because they are outside the port’s operational performance.
Those two wordings sit side by side. The publication page names waiting and time at berth. The note names the start points and the berth departure, and it excludes the post-berth interval. This page does not merge them into a third definition.
The note scores the port, not one terminal. Where a port has several container terminals, possibly with different operators and different productivity, the CPPI reflects the aggregate of the port calls included for that port. The underlying calls come from the S&P Global Market Intelligence Port Performance Program. The note says that program includes 10 of the largest global liner companies, representing approximately 75-80 percent of global container fleet capacity. Ports with fewer than 24 container calls a year in the dataset are excluded.
Rule of thumb: If a port records under 24 container calls in a year in the CPPI dataset, the 10 June 2026 note excludes it from the index. A missing row is a coverage limit. The shipment’s own times stay on the booking papers.
From the 2024 edition onward, the note says the CPPI uses a standardized multi-year series from 2020, with 2024 as the reference year. In earlier editions, scores were normalized to a mean of zero within each year, which the note says made year-to-year comparisons across editions unreliable. A positive value means above-average performance relative to the 2024 reference distribution. The note also says the index does not, by itself, explain why a port time is long. Causes inside the port include productivity, equipment and coordination. Causes outside it include vessel arrival patterns, tidal constraints and labor regulations.
The fuel-consumption weights scale the index. The note says the index weight suggests that recovering an additional hour of port time at sea is 2.57 times more costly for a ship over 13,500 TEU than for a ship in the 1,501-5,000 TEU band. The 2.57 figure is that fuel-consumption weight.
LPI 2.0 uses a different ship clock. Turnaround time at ports is the time container ships call at a port, excluding waiting time at anchorage outside the port vicinity. CPPI includes arrival at anchorage or the pilot station. The two ship clocks are printed with different start rules. This page does not average them.
LPI 2.0 itself is 21 indicators: 6 core indicators, one connectivity indicator and one time indicator for maritime, aviation and postal logistics, and 15 supplementary indicators. The report draws on tracking from maritime, aviation and postal operators, which it says together account for over 80 percent of global goods trade. Data on all 21 indicators are published for 2023 and 2024. Each indicator is reported at the economy level. Cost is not directly measured. Two maritime connectivity indicators, the number of direct services and the number of alliances, highlight competition, and the report says higher degrees of that competition tend to be associated with lower logistics costs.
UNCTAD’s Liner Shipping Connectivity Index (LSCI) is the scheduled-network clock. The UNCTAD indicator page says it can be calculated at country level and at port level, and that it is a proxy for accessibility to global trade through the shipping network. The World Bank metadata file for the series lists six components: scheduled ship calls per week, deployed annual capacity in TEU, the number of regular liner services, the number of liner companies, the size of the largest ship, and the number of other countries reached by a direct service. A direct service may include other stops. The container does not require transshipment.
The same metadata file prints two bases. The definition says the index is set at 100 for the maximum country value in the first quarter of 2006, which was China. The methodology section says each component is divided by the Q1 2023 average, the six components are averaged, and that average is multiplied by 100, so the average LSCI is 100 in Q1 2023. It says this scaling was applied in March 2024 to the whole series from Q1 2006. The file title also says “Average Q1 2023 = 100”. Both bases are printed. They are not averaged here into a third base. The file’s measurement unit line says “Score 0-10000”. The time coverage line runs from 2006-Q1 to 2024-Q1.
Common mistake: Reading a CPPI result as the time for one terminal, one container, or one invoice. The note scores the port’s included ship calls. Your quotes and shipment records are the papers that name what your route included, what it cost and how long it took.
Takeaway: Use the CPPI for the ship’s time until it leaves the berth, LPI 2.0 for economy-level container times, and the LSCI for the scheduled network. Use the booking for the price.
Why smaller exporters feel it more
The World Trade Report 2015, section D, states the fixed-cost point directly. SMEs account for a relatively small share of international trade because there are fixed costs to enter a foreign market that impinge particularly on the profits of small firms. Firms decide whether to enter an export market before they decide how much to export. The same section says burdensome trade procedures, customs and trade regulation are often mentioned as major obstacles to SME export participation, and that a World Bank Enterprise Survey tabulation shows the highest share of firms calling customs and trade regulations a major or very severe obstacle are SMEs.
The same pages then review firm-level studies that do not line up on one size effect. They are printed here as the report prints them.
| Study, as cited in World Trade Report 2015 | What that section says |
|---|---|
| Li and Wilson 2009, Asian firms | SMEs, defined there as firms with fewer than 100 employees, benefit mainly from the “soft” part of trade facilitation, identified as a more transparent and predictable policy. Large firms benefit more from transport and information-technology infrastructure. |
| Hoekman and Shepherd 2013 | Firms of all sizes benefit from a reduction in the average time to export, and the effect is independent of firm size. |
| Hyoungmin and Piermartini 2015 | When all surveyed firms are included, micro, small and medium-sized firms profit more than large firms from lower time to export, and smaller firms are more likely to export and to raise their export shares more than large firms. |
| Espitia and coauthors 2015, Colombian agriculture | Lower domestic transport costs to the port particularly benefit small firms. Exporting firms tend to be smaller where transport costs to the port are low. |
The section’s own summary is that firm-level studies support small firms benefiting from trade facilitation as well as large firms, and that some aspects can benefit small firms more than large firms. The Li and Wilson row is the one that assigns transport infrastructure gains mainly to large firms. This page does not average the four rows into a single percentage.
Unpredictable time is the second channel, and the LPI 2.0 report states it for firms in general. Delays are most unpredictable when goods are stationary, in ports and transshipment facilities, and more so for imports than for exports. Dispersion increases with the median dwell time and is, on average, 36 percent higher for imports than for exports (figure 2.7). Unpredictability disrupts production planning, increases inventory needs, and reduces competitiveness in just-in-time and perishable markets. For firms in global value chains, the report says predictability is often as important as speed. Tracking goods helps firms hedge some of the consequences of unreliability.
The executive summary says some economies have import dwell of a few days, comparable to high-income economies, while others exceed 10 days, and it attributes longer dwell to weak customs automation, inefficient risk management, and cumbersome inspection and clearance. Chapter 2 uses a different cut for the long tail: many economies have exceptionally long container import dwell, more than two weeks, and most of those economies are in Sub-Saharan Africa and in the Middle East and North Africa. Both sentences are the report’s. They are not averaged here.
Figure 2.8 prints three 2024 distributions, which are also left as three rows:
| Port, 2024, figure 2.8 | Leave within 1 day | Half the containers | The slower quarter |
|---|---|---|---|
| Le Havre, France | 10 percent leave in 1 day or less | 50 percent leave in 3.8 days or less | 25 percent take more than 5.9 days |
| Apapa, Nigeria | 0.2 percent leave in 1 day or less | 50 percent leave in 13.4 days or less | 25 percent take more than 22.1 days |
| Algiers, Algeria | 0.1 percent leave in 1 day or less | 50 percent leave in 17.6 days | 25 percent take more than 26.3 days |
Connectivity is a third channel, stated as an association. More connected economies can reach a larger number of trade partners directly and tend to face lower trade costs. The report says direct connections matter because they remove significant delays associated with transshipment, and that the hub-and-spoke pattern has proved cost-effective for carriers while giving small and peripheral economies access to global routes. In 2024, several small island states averaged more than two transshipments: São Tomé and Príncipe 2.6, Anguilla 2.4, the Comoros 2.4, and St. Kitts and Nevis 2.2.
The report also cites two older results with their own scopes. An extra day of inland transit time reduces exports by 7 percent on average (Freund and Rocha 2010). For manufactured goods, the likelihood that a country exports to the United States falls by 1.5 percent for each extra day of shipping time, and by 1 percent for all goods (Hummels 2001; manufactures are SITC categories 7 and 8). Those are the cited studies’ results, not a new LPI 2.0 regression, and they are not a forecast for one firm.
An older port-efficiency estimate is Clark, Dollar and Micco, NBER Working Paper 10353. The abstract says improving port efficiency from the 25th to the 75th percentile reduces shipping costs by 12 percent, and that bad ports are equivalent to being 60 percent farther away for the average country. The introduction says the same percentile move reduces shipping costs by more than 12 percent, and that the result is robust to different definitions of port efficiency and to different years. Both figures are printed. They are not averaged. The transport-cost series is U.S. Import Waterborne Databank data for 1996, 1998 and 2000, liner services only, with more than 300,000 observations a year. The U.S. Census definition used for import charges is the aggregate cost of all freight, insurance and other charges, excluding U.S. import duties, from alongside the carrier at the foreign port of export to alongside the carrier at the first U.S. port of entry. The abstract separately says that reducing country inefficiencies associated with transport costs, from the 25th to the 75th percentile, implies an increase in bilateral trade of around 25 percent. That 25 percent is a trade result, not the shipping-cost result.
Who books the ship often follows the sales channel. How small manufacturers assign shipping to a distributor or a trading house sets out which party owns the goods and which party ships them.
Takeaway: Fixed entry costs weigh on small firms’ profits, as the 2015 WTO section states. The studies it reviews do not agree on whether a shorter export time helps small firms more than large ones, so both results stay on the page.
What a smaller exporter can check on a specific route
A firm does not reset a terminal’s crane rate. It can put the public definitions next to one booking and leave every unprinted cell blank.
Worked example. For a landlocked destination the report’s comprehensive import time is the sum already stated: coastal-port dwell, plus corridor lead time, plus destination dwell. Substitute days only when a document for this shipment prints them. Until then the total stays blank. The “roughly 18 extra days per transfer” figure is the report’s transshipment average in figure 2.3, not a multiplier to apply to a quote that does not say how many transfers it includes.
| What to put on the sheet | What the saved pages support | This shipment |
|---|---|---|
| Same origin, destination, cargo and volume on every quote | Otherwise the rows are different shipments | |
| Stages included, from pickup through delivery | Export dwell starts when the full container arrives, not when the ship sails | |
| Direct or transshipped, and where | LSCI direct service can stop elsewhere if the box is not transshipped. LPI 2.0 counts transfers and time in them | |
| Cut-off before sailing | Export dwell includes buffer time. The printed example is 48 hours before departure | |
| Documents, and who prepares them | Certificates of origin, processed upstream, are not automated in many economies and may delay containers | |
| Part-load consolidation | U.S. ITA lists freight consolidation among forwarder services. It does not print a rate | |
| Tracking and who handles exceptions | The LPI 2.0 report says visibility helps firms hedge some unreliability | |
| Cargo insurance | U.S. ITA lists cargo insurance among forwarder services. Terms stay on the policy | |
| Days and charges | LPI 2.0 does not directly measure cost. No saved page prints this route | blank |
On the U.S. side, the International Trade Administration says international freight forwarders are agents for exporters and can move cargo dock-to-door. The services it lists are arranging and tracking freight, arranging and tracking inland transportation, preparing shipping and export documents, warehousing, booking cargo space, negotiating freight charges, freight consolidation, and cargo insurance. It says most freight forwarders are also non-vessel-operating common carriers, and that both forwarders and those carriers are licensed and regulated by the Federal Maritime Commission. Customs brokers are a separate role: private parties licensed by U.S. Customs and Border Protection to handle the import entry. The page says there are approximately 11,000 active licensed customs brokers in the United States. Other countries set their own rules. The ITA list is the U.S. list.
Import dwell is a time clock at the buyer’s port. The money lines through the buyer’s door are a different sheet, set out in lining up industrial equipment import costs from the quote to the buyer’s door.
Takeaway: Put the same shipment on every row, copy the stage definitions from the public pages, and leave days and charges blank until the booking prints them.
What public policy can change
The LPI 2.0 report says public policy is fundamental even though private operators provide most logistics services. It groups government action in three areas: enhancing connectivity, reducing time and variability, and strengthening institutions and regulation. On time, it lists risk-based inspection, digitized documents and pre-arrival processing, coordinated border management, and gateway work such as berth extensions, yard expansion, access roads and rail links, port community systems, and truck-appointment systems. It says these gateway policies affect ship turnaround, container dwell and the reliability of gateway operations. It does not print a price change on one shipment after a berth is extended, so that cell stays blank.
The OECD Trade Facilitation Indicators monitor border procedures for goods in 163 economies. The 2025 note says progress since 2022 has occurred in every region. It states an average reduction in border bottlenecks and red tape of 3 to 7 percent across regions, and it prints five regional figures: Asia-Pacific 4.4 percent, Europe and Central Asia 3.1 percent, the Americas 4.4 percent, Middle East and North Africa 4.7 percent, and Sub-Saharan Africa 6.5 percent. Those five figures are the note’s. They are not averaged again here. Domestic and cross-border agency cooperation are the top areas of progress and remain the hardest to improve further. Domestic border agency cooperation was already the leading area of reform in 2022. The note also says gaps remain between legal frameworks and practice, particularly when automating documents and processes.
Takeaway: The published reform lists are what governments can move. A firm can watch them and still has to read the booking for one route.
Where this does not hold
- Landlocked origins and destinations. The report says landlocked developing countries face some of the longest and least predictable import times, with most delays from idling at ports, border controls and inland checkpoints. Corridor lead time and coastal-port dwell dominate the import time. The long times are not explained by corridor distance.
- A port score is an aggregate. The CPPI note covers the port’s included calls, not one terminal inside it. Extended port time can come from vessel arrival patterns, tides or labor rules, which the note calls largely outside the port’s control.
- A direct service can still call at other ports. The LSCI metadata counts that as direct when the container is not transshipped. About half of containers in the LPI 2.0 discussion travel directly. The rest may be transshipped.
- A long export dwell includes buffer time. The 48 hours before departure is the report’s example of time the lines set, not a congestion reading by itself.
- The indicators describe 2023 and 2024. A country row is not a quote for a booking made later.
- Road and rail have no global source in LPI 2.0. Note 13 says the indicators do not cover road and rail logistics because no global source of those movements exists, while container tracking on corridors includes some road transport.
- The CPPI series changed its comparison base. Scores from editions before the 2024 standardization were normalized inside each year. The note says those older cross-edition comparisons are unreliable. The current reference year is 2024.
- The NBER 12 percent and “more than 12 percent” results use U.S. liner import charges for 1996, 1998 and 2000. They are not a rate for a shipment this year.
- The LPI methodology page at lpi.worldbank.org returned HTTP 403 on 18 October 2026, so nothing on this page is taken from it.
Takeaway: Treat the indexes as dated, partial clocks. The route decision uses the booking.
How we researched this
The figures above come from the World Bank CPPI publication page and the 10 June 2026 methodology note, Connecting to Compete 2025 (World Bank citation year 2026), the WTO World Trade Report 2015 section D, the UNCTAD LSCI page, the World Bank LSCI metadata file, the OECD Trade Facilitation Indicators note, NBER Working Paper 10353, and the U.S. International Trade Administration page on customs brokers and freight forwarders. The pages were fetched on 18 October 2026. Where one document prints two figures, both are shown and neither is averaged. Sums that the LPI 2.0 report itself defines are labeled as that definition. Day and price cells with no printed source are left blank. How this publication handles sources is on the editorial policy.
Takeaway: The public pages are the evidence for the clocks. The booking is the evidence for one shipment.
