Forecasting Product Flows
December, 2015 in Issue 4 - 2015
I am just turning 75 years of age. Half of my lifetime ago, I was investigating a method of forecasting the trade in bulk liquid chemicals and vegetable oils.
For this, one needs a starting point and we looked at the flows that we could see around the world, based on our own business as well as what we knew of others. The deep-sea fleet then was around 100 ships of 25,000 tons deadweight (dwt) or above. Now it is 1,575 ships and the world has changed!
At that time, looking at the trade patterns, we found that around 80% of the total tonnage carried by chemical tankers on these long hauls either started or finished their voyages in the US. Much of the remaining 20% was tropical oils from Asia to Europe or phosphoric acid from South Africa to Brazil.
Having ascertained that fact, we obtained and analysed all the US trade in detail to give a starting point. Unfortunately I do not have those old analyses; they would be most interesting today.
My employer (Stolt-Nielsen) was the leading chemical tanker operator with about 25% of the world’s deep-sea fleet at the time, and was financially tied up with British Petroleum (BP). Accordingly, one of the senior Stolt-Nielsen staff visited London to get some advice from BP. He returned in awe of the research capabilities of BP which had a team of over 100 people researching all aspects of their business from exploration to the distribution of one litre cans of oil.
On his return from London, he informed me that we should have no problems in compiling a forecast as BP had told him that anything up to a 100% incorrect forecast was acceptable! This was probably due to the upsets in crude oil pricing caused by OPEC during the 1970s but we grabbed that statement with both hands!
Having researched the history of US imports and exports, we applied various factors in order to come up with a forecast of the flows, and thus the fleet results, over the next five years. The freight market in our sector had been appalling for some years, but we forecast that a return to profit would come in the second-half of 1978. We were spot on!
In 2014 the volumes either starting or finishing their voyages in the US were around 20% of the world volume traded, a considerable contrast from the 80% we found in 1978.
The chemical tanker industry developed very fast from that point onwards. Chemical plants were built in other areas of the world that started to compete with the US. The Canadian West Coast developed major plants based on their natural resources and gave strong competition to the plants on the US East Coast, and especially the US Gulf. Major producers consequently moved into that area to take advantage of the newly available feedstocks.
Next came the Middle East. Massive investments in “world scale” petrochemical plants were announced, based on the cheap feedstocks available from Arabian oil, especially Saudi Arabian oil. The announcements were greeted with a certain amount of scepticism as the construction schedules seemed overly ambitious.
In fact, almost every plant came on-stream ahead of schedule. The units were prefabricated, mainly in Korea, and shipped to the Arabian Gulf to be bolted together like a child’s construction kit. On one plant, only two minor adjustments were needed to complete a massive petrochemical plant ahead of time.
The result has been that, over the last 35 to 40 years, the flow of cargo for chemical tankers has changed dramatically, as have the volumes involved.
Regrettably I do not have the data for the years prior to 2008 for all trade lanes; however, I do have the data for the US-to-Far East trade lane from 1997. The volumes over the years up to 2014 are somewhat erratic as Figure 1 demonstrates.
The Asian Financial Crisis that began in 1997 did have an effect on the volumes from the US to the Far East; however, this effect was countered by a poor soybean harvest in South America which resulted in imports coming from the US instead of South America.
A reduction was seen in 2001, 2006 and 2008 in volumes to the Far East. Following the peak in 2010, there has been a steady slide which appears to be continuing in 2015. The slowdown in the Chinese economy has contributed to some of this, but so has competition from other sources such as the Middle East.
I give this trade lane as an example because this is the one where historic data is available. It is very likely that other trade lanes have similar trends; however, this is restricting the picture to US exports and, of course, one would expect some compensation to show up if access to other trade lanes was available.
The full volumes for the last few years are detailed in Table 1.
Historically, the import and export volumes showed similar trends until 2013, when export volumes dropped while import volumes increased. The export decrease was across the board from every area, while imports increased from most areas but especially from the Far East. The year 2014 has reverted to a similar trend, and 2015 seems to be following this as well.
Diversity of products
Product flows have completely changed now. The inclusion of vegetable oils as IMO class chemical cargoes eliminated most of the older, cheaper tonnage involved, and the volumes produced and carried have increased.
Chemicals are moving in increased volumes and in different directions. New plants are still being built; other, older plants are shutting down; and still other plants are being moved to new locations where cheaper feedstocks are available.
The major changes take place for political or economic reasons and the main production companies have experts in analysing the political situations and forecasting changes. They also have teams of economists who predict economic change. Others evaluate the supply and demand of their own narrow field of the business; yet more experts plan changes in their, again narrow, field of the business. The diversity of products can be seen from Table 2.
While the table demonstrates the diversity of products involved, it only tells a small part of the story. The ‘Misc. chemicals’ category for both imports and exports include up to 100 different products, each requiring a different type of tank, different handling, etc.
Several products or product types (shown in bold print in Table 2) are both imported to and exported from the US. There are many reasons for this, such as trading arbitrage, geographical reasons (the US is a large country) or differences in product make-up within a particular group.
It is apparent that there are more product groupings among exports. The number of different products within the ‘Misc. chemicals’ total is unknown, but the production of more of the chemicals is still taking place in the US, especially for those that are more complex and consumed in smaller volumes; hence the fewer number of product groupings in the imports.
Ship owners do not have the luxury of being able to appoint people to evaluate small areas of the business. They must be aware of the potential changes in the trade for a multitude of different products, trade lanes, clients and geographical areas.
Often this is done by one or two people who have to be ‘experts in everything’. They must evaluate the trade patterns of the future as well as gauge political climates and economic developments, and come up with plans for their companies to spend many millions of dollars on building ships that will last at least 25 years and yet have little fixed business except, in a few cases, for one or two years ahead.
Not many, if any, industries will spend this amount of money on such a thin basis. Most of the major operators who are trading with contracts and on several different trade lanes will have a small research department that will collect data, such as I have shown; analyse it in more detail; and advise the company on what they perceive as the future trade patterns.
These companies are normally divided into Trade Lane Departments that take care of individual trade lanes that are covered by their regular business. These departments are mostly so busy planning the next two months and fixing cargoes, plus negotiating contracts, that they do not have time to study their own field in depth, let alone see the bigger picture.
The Research Department can evaluate future flows, based on plants under construction, GDP changes, political changes, etc. They can advise the Trade Lane Departments of what they find but the results are often too theoretical for the departments to act on in a concrete manner.
The other responsibility of the Research Department is often to advise the ship owner on what the future holds and what type of ship will be required in the future. A ship of, say, 20,000-40,000 dwt will take up to three years to build. They can cost anything from US$30 million up to well over US$60 million.
Trade flow forecasts are rarely accurate more than three years ahead, which means that owners can commit to an investment of up to US$60 million and then when the ship is ready, find that the picture has changed and they have an expensive asset that will last for 25 to 30 years and without the base business planned when construction started.
No wonder many ships are now being ordered by trust funds or other financial institutions which are committing other people’s money to these ventures.