When research ‘facts’ are actually incorrect assumptions, why bother with any of it at all?

It is easy to assume that even the most uninterested hard head among us has at some point wondered ‘what does this life all mean?’. The answer is either a mind-alteringly complex version of ‘Id’ or pretty simple – nobody knows. That doesn’t stop any of us having a guess though. From the Pope to Richard Dawkins, from ISIS to the Dalai Lama, everybody is welcome to a guess. So far, nobody has actually been proved wrong. Before you click somewhere else, I promise from here on to stick to the subject of shipping, however utterly mundane it is (as I wrote that last line I really wished that I hadn’t started with the meaning of life stuff now, but anyway . . . ).

One question that I assume we would all love to know the answer to is how big is the universe? How big is any universe? How big is the Yates Wine Lodge in Watford town centre? All laudable and unanswerable questions. I am going to focus on dry bulk shipping here though, but the point does actually remain relevant for most sectors I think. How big is the dry bulk universe? Well Clarksons says that seaborne trade in 2016 was 4.86bn tonnes. So that’s that. Except, SSY says it was 4.34bn tonnes. OK, so there is more than one opinion on this. Only that the error between the two (assuming either is right of course) is over 500m tonnes. I’ll say that again. The two most respected research departments in shipping have an error of more than half a billion tonnes on how much was shipped in dry bulk carriers last year.

So it would be prudent to check with another research company then. Let’s try somebody relatively new to the party in Affinity. They had a figure of 4.45bn tonnes, so close to SSY’s number. But that was for 2012! The 2016 number is 5.06bn tonnes (based on imports by country). Just to reiterate, it is July 2017 as I write this. So what we are looking at here is history. It has happened. And yet the vast brotherhood of shipping is making plans on what is going to happen next, forgetting the fact that the error in what they are looking at is so vast that it is the equivalent of the total imports of Japan and India combined (if you use Affinity’s figures for their imports). Again, it is worth pausing to understand that the headline number of demand for the dry bulk sector is currently being argued by a factor of 700m tonnes per year for the last calendar year. I am not referring to forecasts here. Like the meaning of life, everyone is allowed a theory. But that is all it is. It is not science. It is barely rooted in any form of fact. Somebody may very well be right, but let’s face it, even if they are it doesn’t really matter very much.

I’m assuming that we all think the same thing about this, right?

Now here is a shipping fact. Global steel production is highly correlated to the BDI. How do we know this? I know it because I read an article on a quasi-e-commerce ship broking website, which is a source of some of the most bizarrely out of date ‘pub chat’ masquerading as research in the field. It said ‘ . . . growth in global steel production is very positive for the dry bulk industry since it boosts the imports of raw materials. Therefore a high correlation exists between the BDI and the global production index’. Hell yeah! Unfortunately the author says that this is proved by a graph showing the BDI vs global steel production for the past four years. 2013 you cannot comment on, 2014 shows steel production up, BDI down, 2015 shows both down, 2016 shows steel production up, BDI down. So let’s pretend that we are not talking about dry bulk shipping for a moment. If I said to you that in the last three years on two occasions one figure had gone up and the other down, reasonably you’d say that proves nothing, but it would make me want to know whether these two sets of data have an inverse relationship.

So on our way to prove two things are highly correlated the evidence produced says that they may be inversely correlated, but more realistically, there is no proof either way. This is sacrilege on the conference circuit, blasphemy in the board room, but factually it is pretty much damn near balls dead on accurate. We’ve looked at total demand and found it to be incredibly inaccurate. We’ve looked at one standard assumption and found it to be somewhat dubious. And on we go. GDP is important, industrial production too. Nobody knows what they are within huge tolerances, and having a graph where two wildly different things are placed in different axis and look somewhat similar is not really proof of anything at all, other than the author has a basic grasp of Excel. A Tradewinds headline recent quoted Martin Stopford boldly stating that 60% of shipping demand comes from Asia. A quick check says that of the global population, exactly 60% resides in Asia. The sensible conclusion is that both of these numbers are totally inaccurate, so who cares?

So let’s take our own guess at a seaborne dry bulk trade number for 2016. If we were to say Affinity’s number is too high, SSY’s too low, it doesn’t really matter anyway as we have 700m tonnes to play with, so why not go for half way between the two? Our number is 4.7bn tonnes. I have worked in the dry bulk market for nearly 20 years, so I know that when companies say that they ship 10m tonnes per year they think that is a significant amount of cargo. It’s a number that gives them swag at the Baltic Chairman’s Drinks. Brokers want to make them appear popular and interesting. To be fair, if you piled up 10m tonnes of coal in one place it’s a lot. Then again, it is 0.2% of our estimated global seaborne trade.

It’s dull, but it’s also interesting, but then again, dull.

In context, the biggest of the big charterers says it ships 100m tonnes, 2.1% of global trade (which is guessed anyway). That company trades and ships huge volumes of third party cargoes. Its own demand (it is a power company) is a lot less. What does a typical charterer look like then? A typical charterer ships something like 500-750,000 tonnes per year, with the giants being far rarer than the minnows. If we assign 625,000 tonnes per year to the average charterer then you come to 7,250 charterers in a typical year. It turns out that dry bulk shipping is far larger, lonelier, more chaotic and certainly more difficult to calculate than you read in the magazines.

Here are some examples for you: a fairly new commodity/shipping company based in Asia ships around 10m tonnes per year, including 60% own cargoes. A chartering manager told me that she did not believe that they would not get to see the four supramax cargoes per year I mentioned to her, which are pine nut meal from Vietnam to Germany for a tiny family company based in Hamburg. Many years ago the head of chartering at a publicly listed New York-based shipping company (you can work it out right?) said to me that there was nothing that he didn’t know about the physical market. This was prior to Chapter XI ‘Mark I’of course. I met with the largest aluminium producer in the Middle East who said that they were rapidly expanding their team to cope with 18m tonnes per year of extra shipments. A friend in London told me that there was ‘nothing going on in West Africa’ and there couldn’t be or he would know about it. He operated a small fleet of Supramaxes on a worldwide basis. The list goes on and on.

I recently read a research article on Linkedin by an analyst at a well respected operating company. Based on ‘Likes’ and ‘Comments’ the article seemed to receive nodding approval from readers. I would plead with them to read it again. To paraphrase (really badly, and while I apologise for that bit, the author should apologise for his role in this), year on year volumes of bauxite for one single month were unchanged. Tonne miles, for the month were up though. He mentioned, among other pointless ‘facts’ that Brazilian exports to China were up to 319,000 for the month. If you annualised this noteworthy total it makes up 0.8% of seaborne trade (maybe more, maybe less, we don’t know to within 700m tonnes or so do we?). ‘These changes to a more ton mile intensive trade are VERY POSITIVE for the Panamax and Cape segments’ (my emphasis). Excuse me while I decide if I am going to laugh or cry. I’ll cry actually.

A pansy having a grizzle.

As an ex-girlfriend once said to me: once trust is gone it never comes back. Can we trust any of the known knowns in shipping? I’m genuinely not sure that we can. Do freight rates and vessel values correlate? Yes, of course. So the price of a five year old Supramax rose by 42% from January 2016 to June 2017. At the same time, the Baltic 10 t/cs index rose 98%.

If you want to see the graph below a bit better (pikey WordPress seems to hate my graphs) then click on the thing that says ‘graph‘ below. What it shows is that despite the assumption that the price of a ship is a function of the earnings, this P/E ratio oscillates wildly. (What you are looking at here is the price of a ship in $m divided by the spot index multiplied by 1,800 to represent five years of earnings). Now one can argue this and that about the methodology, but what you cannot say is that the value is a function of the earnings, otherwise the ratio would be close to the average of the period all of the time. It crosses the average about seven times over this period.


While the true meaning of life remains elusive to us, at least we can say one thing for sure, we may or may not actually be alive. In the case of shipping, a calculator might tell you that so many assumptions do not, cannot and never will hold any credibility. Your eyes will tell you that while nothing that is written here proves¬† that there is no correlation, it certainly makes you sit up and wonder if the reason for the correlation or how it works is as simple as people would have you believe.¬†Please don’t misunderstand me, I have literally no answers. All I do know is that so many of the shipping assumptions, research and accompanying nonsense that we take for granted is no more real than Dolly Parton’s hair. And to think that people only want to talk about her music!