By John Richardson
IT has been a remarkably strong few years for the US polyethylene (PE) business. The shale gas revolution has sent ethane costs plummeting, resulting in the kind of margins that you can see in the above chart. Lower oil prices have made naphtha cracking a lot more competitive of late, but the chart illustrates how US ethane-based PE producers still have a significant edge.
And most of the consensus-view forecasts for the next three years are that US margins will only fall from very, very good to good/average, despite the big build-up in new US capacity from now until 2020.
This view largely depends on a healthy Chinese economy because the US exposure to China. The chart below, from our Supply & Demand database, is a reminder of the extent of this exposure.
By 2020, if nothing goes much wrong with economic growth, we expect China to have a deficit of just over 4m tonnes in linear-low density PE (LLDPE).
The implications of a slowing China
In second place will be Europe at 2.3m tonnes, followed by the developing world ex-China that will have much-smaller deficits. Despite all the talk of the “rising middle classes” in the emerging world in general triggering lots more PE consumption, the numbers tell a different story:
- Africa will need to import slightly less than a 1m tonnes of LLDPE by 2020 and South and Central America around 800,000 tonnes.
- The Asia & Pacific region, which includes Southeast Asia and the Indian subcontinent, doesn’t even feature on this chart because we expect it to be in net surplus by 2020.
It is the same pattern in the other two major grades of PE – high-density PE and low-density PE: The US is hugely exposed to events in China.
A severe economic slowdown in China might result in LLDPE demand growth falling below 0ur base-case expectations. This could translate into import demand in 2020 at less than 4m tonnes. The knock-on impact on the rest of the world of a slower China would also be significant, meaning the potential for lower import requirements in Europe and Africa etc.
Our base case for China’s LLDPE demand is that it will rise from around 10.3m tonnes in 2017 to approximately 12.5m tonnes in 2020. This is built on our assumption that GDP will grow by an average of 6.1% per year during this period, with LLDPE consumption growth at an average of 7% per annum. This means we expect LLDPE will grow at a multiple of around 1.16m times GDP.
But during a confidential internal company presentation in Shanghai last week, a leading economic research body gave a very different view of China. Their base case was for an economic hard landing in 2018 as a result of China’s debt problems. They defined this hard landing as 2018 GDP growth at just 4.5%.
In my view, this feels a little too pessimistic. But it is by no means guaranteed that China will escape a financial crisis, and so a hard landing, as its level of indebtedness is higher than in Japan ahead of its 1990s financial crisis – and in the US just before the sub-prime collapse.
So let’s assume, for argument’s sake, that China’s GDP growth does fall to 4.5% next year. Let’s next assume that it bounces back to 5.9% in 2019 and 5.8% in 2020 – our base case assumptions. This would leave average GDP growth in 2017-2020 at just 5.7%.
This would leave LLDPE demand at around 12.4m tonnes in 2020 – some 100,000 tonnes less than in our base case, using the same multiples over GDP. Not a big difference. But what about the knock-on effect on other regions? And is it reasonable to expect GDP to bounce back so quickly in 2019 and 2020?
There is another aspect to this story and this is of course local supply. In times of economic weakness, Western companies tend to cut back on production as their profitability declines.
Not in China, though. History tells us that central and local government officials often insist that petrochemicals and other plants run harder during times of economic weakness. The reason is to generate more tax revenues, and to guarantee employment downstream through greater volumes of local PE resin. Our base case assumes an average local LLDPE operating rate of 94% in 2017-2020. But what if this were to rise to say 97%? This would further lower the deficit, or import requirement, in 2020.
There is another variable to the 2017-2020 local production story, and that is the speed with which new projects are implemented. Some 10 naphtha cracker projects have been announced in China over the last few months, mainly for start-up from 2019 onwards. But what if some of these projects were accelerated for the sake of the “economic multiplier” benefit that the extra construction activity would deliver to a slowing overall economy? This would exert further reduce LLDPE and other PE imports.
The US-China trading relationship
Right now, the US seems to need China on its side, perhaps because of China’s ability to exert geopolitical pressure on North Korea. And/or maybe there has been a shift in the White House’s attitude towards its trading relationship with China, and towards free trade in general.
Whatever the reason, it feels as if the threat of a US-China trade war has receded for the time being, which could bode well for the US’s ability to export PE to China up until 2020.
But this will of course make no difference to the challenge of distance. Competitors in the Middle East India, Southeast Asia and South Korea etc. are much better geographically positioned to deliver to China in shorter timeframes.
Unless, that is, US companies opt for the “logistics hub” approach – building PE warehouses in say Singapore to serve China, or in China itself. Resin would be delivered to these warehouses first, and then orders would be taken from end-users. This would make delivery times comparable with competitors in SEA and South Korea.
Building these hubs would, however, require extra capital costs – and would run the risk of inventory revaluations. Sudden falls in pricing could leave producers with losses on stocks held in these logistics hubs.
Back in the US, port congestion is a big concern as several new US PE plant come on-stream in a short space of time. PwC warns that the US chemicals industry in general could see $29bn extra in operating costs over the next 10 years because of logistic inefficiencies.
There is also the question of how delays might hit sales volumes. How will US producers that don’t opt for the logistics hub approach get their PE to the China market quick enough to win market share from their overseas competitiveness?
And please make a note of this: The US is starting from a weaker position than a decade ago, as in 2016 its total PE exports to China were 460,000 tonnes compared with a million tonnes in 2006.
Perhaps margins will only fall from very, very good to good/average by 2020, at least on a variable cost basis. But will this be enough to pay all the money back spent on new projects? And will the returns be good enough to satisfy the expectations of shareholders?
As always, I could be wrong. And I again hope this will turn out to be way-too pessimistic. But you do need multiple scenarios in today’s very uncertain environment – much more so than before the end of the Economic Supercycle. So I hope that this helps you start the process of building these scenarios.