VoxEU Column International trade

The rum is gone! The impact of maritime piracy on trade and transport

Despite having disappeared from the news since its peak in 2011, piracy continues to threaten maritime shipping. Combining firm-level customs data and ship position data with information on pirate attacks, this column sheds light on the dimensions along which piracy disrupts trade. It shows that exporting firms respond to piracy by switching from ocean to air shipping, while ships re-route in order to avoid affected regions. Despite these adjustments, total exports along affected routes decline.

Maritime piracy has long haunted both global shipping and people living near the shore (e.g. Accetture et al. 2020). However, in times of pandemic-induced closures of ports, a blockage of the Suez Canal by the Ever Given and conflicts between rival governments in the Strait for Hormuz and South East Asia (Cosar and Thomas 2021), it does not come to mind as the number one threat to global transport networks. Nevertheless, modern piracy remains a common threat to international merchant shipping (Figure 1). As a result of the 229 incidents in the year 2020, more than a hundred people were held hostage, several of whom were wounded (IMO 2021).

Figure 1 Worldwide piracy incidents per year


Source: Sandkamp et al. (2021). Data from International Maritime Organization.

Besides the danger to the crew, piracy attacks also lead to delays of ships, as well as damages to the vessel and cargo. Shipping firms adapt by rerouting their ships on costly detours (Bendall 2010) or investing in armed guards, electric fencing, razor wire, water cannons, and other weaponry. Shippers also bear implicit costs of piracy such as wage premia and higher insurance payments. All of these costs have an economic dimension and make it more expensive to ship goods, which ultimately affects the welfare of trading countries.

Pirate attacks are not homogenously distributed across the oceans of the world and hit coastal areas of developing countries more frequently than those of developed countries. Figure 2 shows the distribution of pirate attacks in the time span from 2015 to 2020. Most attacks occurred in Western Africa totalling 385, followed by the South China Sea with 344, and the Malacca Strait with 283 attacks. Piracy has grown as a problem particularly fast in the Arabian Sea and along the coast of India in the past decade.

Figure 2 Distribution of piracy attacks from 2015 to 2020


Note: The map shows the total number of piracy incidents from 2015 to 2020 by region. South America is subdivided into three regions.
Source: Sandkamp et al. (2021). Data from International Maritime Organization

Both the frequently hit Malacca Strait and the South China Sea form segments in the key maritime trade route between Asia and Europe. Hence, piracy is not only a threat to the lives of crews at sea, but also to global trade. In our recent empirical analysis (Sandkamp et al. 2021), we investigate how piracy affects exporting firms’ choice of transport mode, ships’ routing decisions and overall exports.

The impact of piracy on exports and the choice of transport mode

In the first part of our analysis, we use Chinese customs data, which provides information on monthly export transactions at the firm-product-destination-country level for the period 2000 to 2006. Crucially, the data also reports the mode of transportation for the transactions. This allows us to match bilateral trade flows with the number of pirate attacks (taken from the International Maritime Organization) on the set of trade routes linking China to the destination country and analyse the effect by transportation mode. 

At the eight-digit product level, we show that piracy reduces Chinese exports to countries that are supplied by affected routes. Figure 3 illustrates the results using 95% confidence intervals. Effects on export quantity are shown in blue. The results indicate that one additional piracy incident along a set of routes linking China to a particular destination continent reduces exports to all counties on that continent by 0.1%. Given an average of 26 incidents per month along routes to Europe, this implies that exports are 2.3% lower than in a world without piracy. As illustrated in Figure 3, this aggregate effect is solely driven by adjustments in ocean trade, with estimated coefficients for air trade being not significantly different from zero.

Figure 3 Point estimates and 95% confidence intervals of the effect of piracy on Chinese export quantity and shipment size (in %)


Note: On average, 26 piracy incidents take place per month on all routes connecting China and Europe.
Source: Sandkamp et al. (2021).

At the firm-transaction level, piracy reduces the number of transactions carried out by ship, while average shipment size increases (Figure 3). Both results are in line with theoretical considerations by Kropf and Sauré (2014). The authors model that an increase in the fixed costs per shipment – such as investments in protection against piracy – reduces the frequency of shipments while increasing the size of the average shipment. In line with their theoretical framework, the shipment size of Chinese firms increases with the number of pirate attacks. We find that an additional piracy incident increases the ocean-going shipment size by 0.13%. For shipments to Europe, this constitutes an increase of 3.4%. In addition, one additional attack reduces the probability of a firm shipping a product by sea along affected routes by 0.02%. Piracy thus induces firms to switch from ocean to air transport.

The impact of pirate attacks on exporting behaviour is shown to be long-lasting most notably for small firms that might be hit harder by higher insurance costs. In addition, goods with low unit values are on average more strongly affected than those with high unit values.

The impact on ship behaviour

In a supplemental analysis, we match high frequency positions data of large container ships (e.g. Heiland et al. 2020) with geocoded pirate attacks in the time period 2015 to 2020. Drawing on ships’ geographic position and cruising speed allows us to better understand the mechanisms behind reductions in trade at sea. 

Estimated coefficients from regressing the number of ship positions in a given region on the number of piracy incidents indicate that container ships avoid regions recently subject to pirate attacks. Given an estimated coefficient of -0.005 in the most rigorous fixed effects estimation, our results imply that 26 pirate attacks lead to a reduction in the number of ship positions by 12%. This lends support to the hypothesis that ships perform detours following an uptick of pirate activity. It is also possible that the overall number of ships leaving ports to travel to a specific destination declines. In addition, there is tentative evidence that ships passing through affected regions increase cruising speed. By increasing travel-time and fuel consumption, both adjustment mechanisms raise transportation costs and partly explain the overall fall in trade volumes.


Overall, the empirical analysis reveals that piracy negatively affects trade along several dimensions. Exporting firms reduce the frequency of shipments by vessel and switch transportation mode from ship to plane, although the average size of remaining shipments increases. Container ships avoid regions prone to pirate attacks by re-routing and also increase cruising speed, which both increases transport costs. Overall, piracy reduces Chinese exports along affected shipping routes (2.3% for exports to Europe).

Taken together with the dangers piracy poses for the crew of targeted ships, the trade-dampening effects of piracy imply the need for governments to deal with the problem. Increased naval presence would be an obvious short-term fix. In the long run, the improvement of living conditions in countries from which pirates operate might help by eliminating the need for individuals to turn to criminal activity in order to feed themselves and their families.

Even if maritime piracy continues to decline, the results presented in this column may also be relevant to more recent threats facing maritime shipping. In particular, terrorist attacks along the Suez Canal or military assaults by rival governments in the Strait of Hormuz (Cosar and Thomas 2021) may affect shipment costs and uncertainty in ways similar to piracy. Knowing about the potential distortions such conflicts may generate can help policy makers to minimize their impact on the global transport network. 


Accetturo, A, M Cascarano, and G de Blasio (2020), “Long-run consequences of the pirate attacks on the coasts of Italy”, VoxEU.org, 15 April.

Cosar, K and B Thomas (2021), “Disruption of seaborne trade in South East Asia: A quantitative analysis”, VoxEU.org, 04 January.

Bendall, H B (2010), “Cost of piracy: A comparative voyage approach”, Maritime Economics and Logistics, 12(2): 178–195.

Heiland, I, A Moxnes, K H Ulltveit-Moe and Y Zi (2020), “Trade from space: Shipping networks and the global implications of local shocks”, VoxEU.org, 07 January.

IMO (2021), “Reports on acts of piracy and armed robbery against ships”, MSC.4 265.

Kropf, A and P Sauré (2014), “Fixed costs per shipment”, Journal of International Economics 92(1): 166–184.

Sandkamp, A, V Stamer, and S Yang (2021). “Where has the rum gone? The impact of maritime piracy on trade and transport”, Review of World Economics, forthcoming.

3,150 Reads