Columna De Expertos Obstacles of Ocean Freight Prediction: Variability, Uncertainty, and Data-Securing Difficulty

The maritime industry is one of the important industries that lead the flow of the world economy. The core of this industry is the “ocean freight rate,” and it is important to understand this concept when identifying the trends of the entire industry. Simply put, ocean freight refers to the rate one receives for transporting merchandise with a ship. We use many goods in our daily lives that are shipped from different parts of the world by ships, and the transporting fees are reflected in the pricing of goods.

It is very important to predict ocean freight rates because their fluctuation has a significant influence on the global economy and trade. If the prediction is accurate, not only trade companies and shippers but also consumers can benefit from it. On the other hand, if the prediction is inaccurate, unnecessary costs are incurred or opportunity costs can disappear. Therefore, accurately predicting the trends of ocean freight rates and coming up with strategies is not only significant in the industry but also in the entire global trade.

Then how are ocean freight rates formed? Many factors influence the freight, but the main players are supply and demand. However, complex characteristics and patterns also play a role in the back of supply and demand. The maritime industry follows a certain cycle: trough, recovery, and collapse. A simple description of this cycle is as follows. A carrier adjusts the supply of a ship based on the demand and ocean freight level. If demand increases, the carrier can increase ship orders. In contrast, if demand decreases, the carrier can select between blank sailing, low steaming, and even scrapping. Through this adjustment cycle, the freight is finally formed in the balance point between supply and demand. If one acknowledges the pattern of this cycle, they believe they can easily predict ocean freight. However, there are many obstacles in predicting the freight. This column deals with these difficulties.

1. Demand Using various data is necessary to predict ocean freight rates from the perspective of demand. Some examples of data are cargo volumes, Purchasing Managers' Index (PMI), Gross domestic product (GDP), the composite leading indicator (CLI), etc. However, the accuracy of prediction is heavily reliant on data reliability. This is where the problem lies. It is impossible to secure the current data for cargo volumes. For example, the ocean freight data can be secured until September 2023. However, cargo volume data can be secured only until July 2023. This data vacancy of 2 months becomes a big variable when predicting the rates.

In fact, Container Trades Statistics and Clarksons Shipping Intelligence Network, the world-renowned providers of ocean data, and many other institutions have a 2-month data vacancy from the current time point when it comes to cargo volumes. The reason is crystal clear. This is because data accuracy and integrity are superior to the volumes. It takes a significant amount of time to collect, verify, and process data from carriers, ports, and other relevant institutions, so it takes about 2 months.

The 2 months might seem meaningless, but they play a significant variable in the real prediction of ocean freight. This time gap significantly affects ocean freight prediction which is updated every week and negatively influences the accuracy and reliability of the prediction.

2. Supply Calculating supply is very important when predicting ocean freight rates, but it is difficult. The main problem is that it is not easy to find precise data on routes. The Shanghai Containerized Freight Index (SCFI) is a representative index of container freight rates based on Shanghai. It includes freight rates of the major 15 routes centered on Shanghai. Some examples of important routes are China-USWC, China-North Europe, China-Mediterranean, and China-USEC. Therefore, detailed data on demand and supply of those routes are necessary when accurately predicting freight rates for each route.

It is relatively easy to secure data related to cargo volumes, but it is not easy for supply information for each route. In particular, it is extremely difficult to secure detailed supply data per route. Clarksons Shipping Intelligence Network and Drewy, the two main providers of ocean information, focus on information between continents (Asia-North America, Asia-Europe, etc.) but they have limitations in that they cannot provide detailed data per route. Of course, Clarksons offers data, such as shipping capacity, delivery, contract amount, and scrapping, but they are not separated for each route. Therefore, they are helpful for understanding the big flow of the world ocean industry but are not enough for predicting detailed rates of a specific route. Moreover, access to information regarding blank sailings, slow steaming, GRI strategies, etc. is limited.

Due to these limitations, it is hard to calculate and predict the supply of each route. To address this, I am currently cooperating with Samsung SDS on a project to predict ocean freight rates for 3 years. In this process, we use schedule reliability data as an important index. Adding to this, we calculate the supply of each route based on carriers’ accurate schedules and vessel-input information. What’s more, we are developing an algorithm to predict and analyze the possibility of blank sailing following the fluctuations of ocean freight rates for accurate calculation of supply for each route.

3. Global Issues-Geopolitical Conflicts and Climate Change When deep-diving into the fluctuation of the maritime industry, one can easily recognize it is due to the interconnectivity of various factors. In particular, ocean freight rates are heavily influenced by various external factors that are difficult to predict, including global politics, economic situations, and climate change. These factors interact with one another in a complex way and make it more difficult to predict the maritime industry.

An unstable situation is one of the main factors that heavily influence the global logistics and maritime industry. This instability can bring about changes in many factors, such as trade routes, transportation volumes, and logistics fees caused by political and military conflicts.

A representative case is the conflict between Russia and Ukraine. This conflict disrupted the supply flow of key energy and agricultural goods in the world and resulted in a big fluctuation in global trade and ocean freight rates. The recent Israel war is another issue that needs continuous attention. Carriers who provide services in the ocean near Israel might have to change or adjust their service routes if the war gets prolonged. This could then lead to an increase in ocean freight rates. This instability of the situation makes it more difficult to predict the rates. The freight is dependent upon the balance between supply and demand. Hence, trade discontinuation, route changes, supply chain changes, etc. caused by unstable situations heighten the complexity of the prediction.

This shift can be seen in route-specific volume data. Looking at route-specific data from Container Trades Statistics (CTS), container traffic from China to the US West Coast was 586,715 TEUs in December 2021, but dropped to 458,765 TEUs in December 2022 following the war between Russia and Ukraine, a 21.81% decrease. Container traffic from China to the Middle East fell from 392,293 TEUs in December 2021 to 268,281 TEUs in December 2022, a decline of 31.70%. Container traffic from China to North Europe decreased from 764,176 TEUs in December 2021 to 595,281 TEUs in December 2022, a decrease of 22.20%. However, container traffic from China to the Middle East increased from 204,695 TEUs in December 2021 to 257,693 TEUs in December 2022, an increase of 25.89%. Container traffic from China to South America decreased from 115,623 TEUs in December 2021 to 110,779 TEUs in December 2022, a slight decrease of 4.19%.

In general, container cargo volumes increase upward. However, the trends of container cargo volumes are changing after the Russia-Ukraine war in major routes. These changes decrease the accuracy of ocean freight prediction. Therefore, data analyses of more precise and overall cargo volume data are important. The prediction model needs improvement considering the changes in the entire world trade flow.

Another case can be climate change. One representative issue is the drought in the Panama Canal. According to the Panama Canal Authority, the level of Gatun Lake marked 79.65 feet, a drop of 6.8% compared to the 5-year average. Due to the problem in the operation of the Panama Canal, draft limitation started, reducing the number of vessels that pass through the Canal from 36 to 32. Due to this draft limitation, container vessels should reduce their capacity. They may select the Strait of Magellan as an alternative route, but given that the containership’s vessel speed is 10 knots, it is inefficient since the sail time increases by about 33 days compared to when using the Panama Canal. The drought in the Panama Canal led to an increase in ocean freight rates; the rate from Shanghai to USEC recorded 2,869 dollars per 1FEU in the second week of September. It is an increase of 43% compared to that of the last week (2,010 dollars) of March. 4. Seasonality Another factor that complicates the prediction of ocean freight rates is the fluctuation of seasonality. The maritime industry is seasonal: its characteristics and patterns appear in a certain period. The third quarter is a representative example of this seasonality: freight changes of regular lines are shown during this period every year. In this period, Thanksgiving and Christmas in the US (4th Thursday in November and December), cargo volumes skyrocket due to the securing of large-scale inventory from retailers. As a result, ocean freight rates show an increasing trend. In addition, another factor that has a huge influence on this seasonality is the National Day of China (Oct 1, 7-day holiday). Cargo volumes plummet as many companies and factories go on a vacation in China during the National Day. To address this, many carriers choose blank sailings, adjusting supply. As a result, the freight often increases during the National Day.

However, an exceptional case occurred in the third quarter of this year. The SCFI recorded 885 points in the last week of September, dropping under the 900 points mark for the first time since May 2020. This is related to the overall downturn of the world economy, and the main reason is the large-scale decrease in cargo volumes. Unlike the consistent seasonality pattern in the past, the current situation is a huge obstacle to the prediction of ocean freight rates. The 2-month delay of cargo volume data negatively affects the precision of predicting ocean freight rates.

5. Conclusion This column discussed various obstacles when predicting ocean freight rates. The key difficulty of the prediction lies in the obstacle of securing data and huge variability. At the current moment, it is difficult to secure cargo volume data, and the changes in the global supply chain make it even more complex. In addition, unstable situations and fluctuations of seasonality are also factors that hinder the prediction of ocean freight rates.

Many factors are required to address this problem: analysis of changes in the global cargo volume data, a prediction model based on diverse variables, discussions with experts, and so forth. In addition, it is important to observe seasonality and changes in cargo volume in real time, adjust strategies according to each situation, and respond. This way, the maritime industry can increase prediction accuracy and decrease variability risks. Moreover, technical approaches, such as Big Data, Machine Learning, Deep Learning, and simulation will help strengthen the efficiency and accuracy of prediction.

In particular, the project of predicting ocean freight rates that I have been working on with Samsung SDS for 3 years is a key effort to address this problem. This project suggests the method of predicting rates and may be of great help for carriers and shippers when setting short- and mid-to-long-term strategies. # Reference [1] Stopford, M. (2008). Maritime economics 3e. Routledge.
[2] Container Trades Statistics
[3] Clarksons Shipping Intelligence Network

▶ This content is protected by the Copyright Act, and the copyright is held by the contributor.
▶ This content is prohibited from secondary processing and commercial use without prior consent.