Infrastructure and Trade Costs

Both the quantity of infrastructure investment and the quality of infrastructure services influence trade performance (see e.g., Limao and Venables 2001; Clark et al. 2004). This occurs through infrastructure's impacts on monetary transaction costs, loss, damage and spoilage to goods in transit, and timeliness of delivery, among other factors.

Nordas and Piermartini (2004) delineate four dimensions of the relationship between infrastructure and trade transaction costs:

1. Direct monetary outlays for delivering traded goods are partly determined by the quality of infrastructure and the cost and quality of related services.
2. Timeliness, even more than freight rates, is likely to be influenced by geography and infrastructure.
3. Risk of damage, losses, or larger insurance costs is higher when infrastructure is of poor quality.
4. Lack of access to a good transport or telecommunication service can have a high opportunity cost, restricting market access and limiting the likelihood of participating fully in the benefits of trade.

An important component of transportation costs is the time cost involved. This is particularly critical for perishable or other time-sensitive goods. Hummels (2001) found that the time cost of one day in transit for US imports is equivalent to an ad valorem tariff rate of 0.8%, implying the equivalent of a 16% tariff on an average ocean shipment of 20 days. Clearly, improvements in infrastructure services that reduce delays in ports, border crossing procedures, or transit times will influence a country's propensity to trade. Developments in containerization and intermodal transport networks contribute to quicker delivery times and the growth in air shipments.

With the value of timeliness in delivery rising in recent decades, congestion is becoming increasingly costly. When growth is very rapid, congestion results as the increase in traffic induced by the economic growth outpaces the expansion of transportation infrastructure services. Ma and Zhang (forthcoming 2009) find this to be the situation in the People's Republic of China (PRC). Sea port congestion there results from the long neglect of access transport and port facilities infrastructure. Six percent of the world's rail lines struggle to move one-fourth of the world's rail freight turnover, and only 2% of the country's highway network is expressways.

Congestion has been rising, notably at the port of Shanghai, as overloading of the physical infrastructure is compounded by a lack of collaboration among different stakeholders at the port in achieving greater levels of supply chain efficiency. This reinforces the drive to increase port and modal competition for greater gains in efficiency by increasing both hard and, increasingly, soft infrastructure. In terms of soft infrastructure, reliability of trade facilitation and administrative procedures at customs are crucial, including rationalization of the customs transit system in order to reduce inspection time and simplify declarations and the documentation process. Meanwhile, Shanghai's congestion is reducing its competitiveness relative to nearby ports in neighboring economies, endangering its status as a hub and premier gateway to international markets and suppliers. In recent years, transshipped containers from Shanghai via Hong Kong, China have accounted for as much as 20% of the total container throughput of Shanghai.

The limited extent of infrastructure connections to western regions of the PRC results in high trade costs for inland regions and impedes regionally balanced growth. As land and labor costs rise near coasts, investors are looking to locate production facilities farther inland. However, they are hampered by poor infrastructure. This has led to a shift of emphasis in infrastructure policy that gives greater weight to hinterland access. In particular, railway construction is crucial for inland provinces, where a greater share of production is of bulk commodities.

The composition of freight charges can vary significantly across countries and commodity categories. De (forthcoming 2009a) finds that the share of Asia's total freight charges accounted for by inland freight may be less than that by ocean freight, but is frequently greater. The actual balance depends on country characteristics, suggesting an inland focus for trade-related infrastructure priorities in those countries where the inland share is greater and there is a role for regional cooperation in incorporating landlocked countries into international trade patterns. From 2000 to 2005, transport costs became relatively higher and shipping distance relatively lower, and a 10% rise in transport costs (expressed as an ad valorem tax equivalent) is found to lower Asia's trade by about 3%–4% from what it would otherwise be. When trade is differentiated by commodity groups, the weight to value ratio is found to be the major determinant of transport cost, suggesting that road, rail, and sea may be the increasing order of modal preference for transporting heavier cargos in Asia.

Hummels and Skiba (2004) similarly found that a 10% increase in the ratio of product weight to value leads to a 4% increase in ad valorem shipping costs, reflecting the demand for higher value cargos. From the consumer's point of view, higher shipping costs can reflect a smaller ad valorem charge in the final price paid, so the consumer is more likely to use more expensive modes of shipping when the impact on the delivered price is smaller.

The relative weights of different categories of trade costs are often surprising. In the same study De notes that in 2005 the ocean freight rate for importing a container to India was about two thirds greater than for exporting. At the same time, in the PRC ocean freight for importing a container from six Asian countries was far lower than for exporting. Auxiliary shipping charges (documentation fees, container handling charges, government taxes and levies, etc.) may account for much of this difference and are sometimes greater than the ocean freight charges, particularly where shipments experience congestion at ports or borders. On average, auxiliary shipping charges outweigh terminal handling charges across countries and commodities in Asia, with variation in such charges contributing significantly to variations in trade costs. This highlights one crucial area of soft infrastructure's potential contributions to lowering trade costs.

Domestic infrastructure behind the border can have as much effect on the length and variability of time-to-market as freight services between countries. This is especially true in large or landlocked countries, and the proliferation of inland dry ports has evolved partly in response to this problem. Limao and Venables (2001) found that domestic infrastructure explains about 40% of transport costs for coastal countries, while domestic and transit country infrastructure together account for an estimated 60% of transport costs for landlocked countries. Furthermore, land transport is about seven times more costly than sea transport over similar distances, and estimates of the elasticity of trade flows with respect to transport costs range from -2 to -3.5, suggesting that lowering a country's trade costs by 10% through infrastructure development could increase its exports by over 20%.

De (forthcoming 2009b) finds that inland transport cost is the major component, accounting for about 88%, of overall trade transportation costs in South Asia. Inland costs are very high across South Asian countries, except in Sri Lanka, and vary across goods and countries, being even higher when countries are landlocked. Land border crossings are overcrowded, and greater policy attention to efficiency concerns could easily reduce delays and monetary costs. Complex border-crossing requirements in trade expand possibilities for corruption and have encouraged sharp growth in informal trade. The magnitude of border effects in South Asia argues strongly for improvements in soft infrastructure, complemented by inland transportation infrastructure, to raise the competitiveness of the sub-region's exporters.

Download this Paper [ PDF 177.1KB| 21 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.