Tugrul Vehbi, Serdar Sengul, Daniel Christen, Lucio D’Aguanno and Tom Wise
Shipping costs have increased sharply since the onset of the pandemic, to a magnitude perhaps only a few would have predicted. In this post, we examine the likely drivers and impact of this increase. We argue that (i) both demand and supply factors are responsible for these developments with the former playing a relatively bigger role historically; (ii) shipping costs feed through to consumer prices with a lag; and (iii) therefore, we may expect to see further price pressures in some advanced economies (eg the US and the euro area) from recent surges in shipping rates.
Ocean freight rates have risen very sharply since the second half of 2020 and have reached historically high levels, as can be seen in Chart 1, which shows five key indices of freight rates. The most eye-catching rises in rates have occurred on container ships (or ‘box ships’) such as the ‘Freightos Baltic Container Index’, an index of spot rates on a weighted average of 12 trade lanes. The rise in this index has been driven primarily by sharp rises in rates on routes from Asia to North America, and also on routes from Asia to Northern Europe/Mediterranean.
Chart 1: Global freight rate indices
Source: Refinitiv Eikon from LSEG.
Global excess demand for goods is predominantly driving the shipping rates
Behind this increase in shipping costs lies a mismatch between a strong global demand for goods and several supply constraints in maritime transport. On the one hand, the goods-intensive global recovery has seen a sharp pickup in manufacturing activity, increasing international trade in intermediate inputs and demand for container shipments. In terms of supply, the asynchronous economic recovery had seen empty containers left in several ports in Northern America and Europe, creating a shortage of containers available for export from Asia. At the same time, a series of Covid-related disruptions at Chinese ports created delays and eventually congestion at ports in Europe and the United States, with vessels arriving with a delay of over 7 days on average in September, an excess of 3.5 days relative to the 2016–19 average. This has boosted freight rates to multi-year highs, particularly on shipping routes from Asia to North America and Europe. And until recently, the surge in global oil and fuel prices have further increased the shipping rates.
To gauge the relative contributions of these demand and supply factors to freight rate movements, we have adopted a three-variable structural vector autoregressive (SVAR) model following Attinasi et al (2021). The model comprises monthly data of container shipments on 15 major trade lanes, shipping prices as measured by the Harper Petersen Charter Rates Index (Harpex), and oil prices as measured by Brent futures from September 2011 to August 2021. We have identified the shocks to demand, supply and oil with sign restrictions. A positive demand shock should lead to rising shipping prices and quantities, whereas a supply shock would move them in opposite directions. Finally, the oil price shock should lead to an increase in oil and shipping prices, and a fall in the quantity of shipments.
Chart 2 shows a historical decomposition of freight rates, using the SVAR methodology outlined above. Historical decompositions are useful for explaining how much a given shock identified by the model explains the historically observed fluctuations in the model variables. In line with similar findings in the literature, we found that demand shocks (blue bars) historically dominate supply (orange bars) and oil (brown bars) shocks as drivers of freight rate movements, and the recent rise in freight rates is no different. That said, supply factors somewhat offset the demand drivers around 2020 Q2, as can be seen in Chart 2, from the orange bars pulling the monthly change well below zero. From the beginning of 2021, however, both demand and supply factors have contributed to the recent pickup in shipping costs.
Chart 2: Historical decomposition of freight rates
* In deviation from its deterministic path, ie the path shipping costs would have taken if no shock occurred since the starting point.
Source: Authors’ calculations.
What about inflation?
The ultimate question though is about what this surge in shipping rates might mean for consumer prices. Quantifying this pass-through is not straightforward, as shipping costs are generally not captured in global goods trade price indices. Transportation costs often are borne by importers, who can pass them onto consumers. Inflationary pressures depend on the degree of such pass-through. Herriford et al (2016) estimate it with a four-variable SVAR model comprising oil prices, as measured by World Texas Intermediate (WTI) spot prices, a nonpetroleum import price index, shipping prices as measured by the Harpex index, and the core PCE price index. Their results suggest that a 15% increase in shipping costs pushes up US core PCE inflation by around 10 basis points in the first year, peaking after 11 months. Most recently, the ECB and OECD found that a 50% annual increase in shipping rates lifts annual US personal consumption expenditure (PCE) and consumer price inflation in OECD countries by 25 basis points respectively.
However, recent surges in shipping rates have been so dramatic that producers/importers may have had to pass the costs on to consumers to a greater extent than in normal times. To reflect recent possible behavioural changes, we re-estimated the pass-through to inflation in the United States and the euro area with a slightly extended version of the Herriford et al (2016) model. Specifically, we augmented the original specification with several additional control variables, including food and metal prices as well as the movements in exchange rates, to isolate the contribution of exogenous changes in shipping rates from demand-related endogenous factors. We estimated the model with monthly data from January 2001 to August 2021, using three lags for each variable. We have also done a number of robustness checks including on the lag order, the recursive order, and with a more limited data set excluding the recent surge.
Overall, we found that a 1 standard deviation monthly increase in shipping rates pushes up the level of US PCE and euro-area HICP by around 0.07% and 0.05% respectively at peak, approximately one year after the initial shock (Chart 3). This is consistent with slightly larger impacts than the studies mentioned above.
Chart 3: Response of US PCE and EA HICP to a 1 standard deviation increase in shipping costs as identified in the SVAR
Source: Authors’ calculations.
How should we interpret these results?
Soaring freight costs are likely to push up global inflation further should they continue to remain high. However, it is difficult to be precise about the full impact this might have for consumer goods prices. Intuitively, globally rising costs could be forcing producers to pass these costs to consumers to a greater extent than usual. However, the inflationary pressures from shipping costs are likely to have been limited so far. That is because firms are still likely to absorb some of the rise in shipping costs and shipping only accounts for a relatively small share of the total cost of manufactured goods. Nevertheless, we might continue to feel the inflationary pressures from elevated shipping rates until at least mid-2022: the impact of rising shipping rates on inflation occurs with a lag, reflecting the fact that many (typically larger) importers fix rates for the duration of the contract, which are typically over a year in length. This means that for those operating under such contractual rates, rising rates only get reflected at the point of renewal.
Tugrul Vehbi, Serdar Sengul, Daniel Christen, Lucio D’Aguanno and Tom Wise work in the Bank’s Global Analysis Division.
If you want to get in touch, please email us at firstname.lastname@example.org or leave a comment below.
Comments will only appear once approved by a moderator, and are only published where a full name is supplied. Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.