An inexact science

But is forecasting ever accurate enough? “The answer is pretty well no,” says Niels Westberg, founder of Brabazon Fox and a member of the consultancy alliance UK Port Advisers. “It is such a complex game that it is almost impossible to be exactly right. Of course you can be partly right – but that can be misleading.”

As haven master at the Port of Bristol until two years ago, Capt Westberg was heavily involved in masterplanning and making a case for the proposed deepwater container terminal in the river outside Avonmouth docks; he also had experience of working ‘against’ a generally accepted forecast that had been drawn up by MDS Transmodal for the UK’s Department for Transport (DfT). The MDST forecast suggested that, other than existing traffic at Liverpool, deepsea shipping lines would be reluctant to make calls at ports outside the UK’s Greater South East.

Bristol disagreed – and it did so vociferously, because such an assumption could undermine belief, or confidence, in its plans – whether among shipping lines, shippers or investors – and could also affect its chances of gaining planning consent.

“We analysed about a quarter of a million of box moves going to port of entry and post code destination, so we knew our figures were real, and we used that as a basis for challenging the report,” says Capt Westberg.  

“Under the [EU] Habitats Directive, you have to prove need; and underpinning need is forecasting. So the government and its agencies have to look at that. For banks and investors, it is different; their approach is ‘before I lend I have to have proof’. They presume government data is enough proof – it is the only way they can get it.”

Suddenly, the impact of a forecast you don’t agree with becomes clear. Bristol spent three years challenging the DfT report: “It does take time to break free from the shackles of received wisdom. We had to overturn received wisdom, and we won in the end,” he says.

However, Capt Westberg adds: “Yes, you have to prove need. But that is a slightly theoretical hurdle to jump because it is quite different from the reality of commercially winning the trade. People should really know their own business.”

Richard Clarke, global director, ports and marine, at consultant AECOM, uses demand forecasts to generate cash flows for projects but he says: “One certain thing about demand forecasting is that they can be wrong.”

However, he says: “My approach is that unless something catastrophic happens, like someone opens a competing port next door, generally there will be growth – because port business grows all the time. The uncertainty and demand forecasting just varies the date on which things happen. If you think you need an extension of your port in 2020 and you then get a global recession for a few years, OK maybe you don’t need it until 2028.”

Indeed, there is a certain inevitability that there is going to be growth, says Capt Westberg – for example, the UK population is around 60m but there is talk of 70m already. “If the population is growing, then GDP will grow. There are a limited number of big ports and therefore if one is taking the long-horizon view, which one must, then one isn’t going to get too far unstuck, because growth will come.”

AECOM’s Mr Clarke says the same principle could be applied to, for example, Iraq, where a lack of any real history makes cargo forecasting a real challenge. “In Europe, each person uses ‘X’ number of containers of goods, so you can take the population of Iraq and predict that they will get there eventually … but it is impossible to come up with a date, of course.”

Sensitivity testing is always an important part of the process in creating financial models for ports, he says – i.e what happens if you only get half the expected growth rate, and how would that affect the sums? “I get endlessly accused of being a killjoy. People just want to hear what they want to hear – unless you meet the finance manager!" he says.

“We knew we needed six years’ lead time to get consents and build, so the whole model was put together so that they knew the critical data they needed to ‘push the button’,” he says.

“It is interesting meeting different clients; attitudes vary tremendously. Dover was pragmatic and prepared to adjust year to year; others just say ‘go and build it and the trade will come’. The Middle East ports are famous for that. But it is commercial clients, who have to get a return for their money, that are more dependent on forecasting.”

What a forecast does do is set a time framework, “rather than actually predicting that trade in this year will be that”, says Mr Clarke. “It is much more a matter of when trade gets to this point, you will need new facilities or two new cranes; it is predicting the trigger point.”

Forecasting in connection with the extension of an existing port is far easier than trying to make predictions for starting from scratch on a greenfield site, he adds. And always, “some really surprising things” can come along later.

For example, Mr Clarke has a long association with Mauritius – including producing masterplans for Mauritius Port Authority first with Royal Haskoning and later with Halcrow. “When we first built the container terminal, we really struggled to justify it; and now they are in the process of extending it.

“That has nothing to do with trade to Mauritius, but the fact that ships coming down from Suez to Australia are crossing over with ships from Singapore to South Africa, and a certain number of boxes needing to move from one to the other. Almost overnight this transhipment business doubled the throughput of the terminal. That was completely unpredicted, except by MSC, and that sort of thing is incredibly difficult to forecast.

“If you are forecasting wine imports into the UK, really that is a function of how the economy is going and that sort of thing is quite easy to do; but predicting bit structural changes in shipping patterns is much harder.”

Neil Davidson, Drewry’s senior advisor – ports, agrees. “It is all about gathering the data that you can and seeing how much you have to work with. For example, if there’s a steel works next door to the port, as long as the steel works stays open then certain volumes can be expected. But if the steel works could shut, the forecast is zero. It is much more difficult to forecast transhipment; it is easy to change and the decision of one person sitting in an office can change volumes overnight from 500,000 teu to zero.”

In a sense, it isn’t about actual numbers coming out of a forecast, he says. “Maybe the more important thing is to help the client or port to fully understand the drivers behind the forecast and the issues affecting the forecast. It is more about helping to paint a picture and understand the soft side of it, rather than saying that in 2036 the port will have ‘X’ million teu.

“It is more about saying these are the things that will affect your market position, your volumes and your market and these are the strengths and weaknesses – to help ports and also banks to understand the risks and sensitivities.”

The role of the forecaster or consultant is to help clients make their own informed decision, not to make the decision for them, says Mr Davidson. “Ultimately, people have to make their own judgment – for example, the bank whether it is going to lend money or the terminal operator whether it is going to invest in that facility or buy that company.”

Drewry has been asked to forecast everything from cruise passengers to livestock and it can be frustrating not being able to get the data you need to work with, says Neil Davidson. “The other frustration is that the world can turn around and bite you, with an economic crash that no one expected.”

And there is always a limit to what forecasting can do, says Capt Westberg. He highlights an airport policy review drawn up in the UK about two decades back, which missed in its entirety the birth of the low-cost airlines. “And they turned airports on their heads in many ways.”