Market Fragmentation and Inefficiencies in Maritime Shipping

Maritime transportation is critical for the global economy, with about 90% of goods traded by sea and volumes expected to triple by 2050. Despite rising demand, 40% to 50% of total miles are sailed empty (known as ballasting), causing significant economic and environmental costs. The oil transportation market is an ideal setting to study causes and remedies of ballasting, because: (i) oil tankers only carry a single cargo from origin to destination; and (ii) the market is highly fragmented.

Ballasting is inevitable where demand flows are imbalanced, but imbalances are not its only cause. The presence of many independent carriers, their interactions, and their strategic decisions, such as route selection, also influence the overall transportation efficiency. This paper studies the role of market fragmentation in ballasting and highlights that economic and environmental outcomes can be considerably improved by organizing existing resources more efficiently. We capture these inefficiencies both numerically and empirically using a proprietary dataset containing approximately six thousand voyages of oil tankers from 2018 to 2022.

We first formulate and solve a set of integer and stochastic dynamic programs under different market structures and informational assumptions. Using their solutions, we estimate the share of ballasting associated with trade imbalances, market fragmentation, and demand uncertainty. We find that fragmentation is responsible for 15--20% share of ballasting costs, on top of a 70--75% baseline caused by trade imbalances. We further refine our analysis to show that even partial consolidation of the market into shipping pools of 30--40 units leads to a reduction up to 15% in overall ballasting costs, thereby mitigating the economic and environmental costs without excessively increasing the market power of shipowners.

We then analyze the movements of existing operators to understand the mechanisms underpinning the positive effect of consolidation. Using two different empirical strategies, we find that, on average, doubling the size of a fleet is associated with an increase in utilization (full miles/total miles) between 2.2 and 4.4 percentage points. This improvement is mainly achieved through two channels: a coordination-effect, whereby vessels managed by the same entity can serve the same locations at a lower cost; and a network-effect, whereby larger pools serve a more complex network of locations, allowing them to run more integrated operations. In contrast, smaller pools tend to concentrate on fewer routes. Finally, we confirm that there are decreasing marginal returns to efficiency from increasing the fleet size. Consistent with our numerical results, we find that pool sizes between 20--40 vessels suffice to obtain most of the efficiency gains.

The full version of this paper is available at: https://bit.ly/3KPZCB9.