The Ethylene Conundrum – EMN the Value Play

Print Friendly, PDF & Email


Graham Copley / Nick Lipinski / Torey Shepardson
203.901.1629 / 203.989.0412 / 203.989.0411
gcopley@ /nlipinski@/

January 5th, 2015

The Ethylene Conundrum – EMN the Value Play

  • While likely unsustainable, an estimate of the global ethylene cost curve for late December shows us two things: how much money the global industry made at the tail end of 2014, and how tenuous investments in the US could look with more normal costs and pricing (Exhibit 1).
  • If we assume prices/estimates fall in line with lower costs, from a valuation perspective LYB and DOW discount more of the possible downside than WLK. EMN is the obvious long idea as its business needs propylene – cheaper with lower propane – and the stock has valuation support
  • Global prices for ethylene and its co-products are tumbling, with monthly price declines far larger than historic norms. However, outside of US ethane, feedstock costs are falling significantly more quickly creating what is likely to be short term very high profitability.
  • The apparent negative costs below are on facilities cracking propane in the US, though there are a large number of naphtha units globally where variable costs (excluding direct labor and shipping) are close to, or below zero. Ethane based ethylene units have no co-product credits except fuel and cannot exhibit negative cash costs, even if ethane is free.
  • Most raw material and co-product price scenarios have US ethylene profitable (expensive co-products and low crude would change this). But it is hard to find conditions that support 2015 and 2016 consensus estimates for LYB, DOW and WLK, especially WLK which does not benefit if European margins are better. Detailed cost curve analysis supports our October 2014 view that ethane based ethylene margins in the US could fall by 15 cents per pound in 2015 versus 2014 (41%)
  • Those pushing ahead with major investments in the US are betting on a cost/price scenario which now looks very much like that; a bet! We discuss scenarios that would still make these good ideas, but we also highlight a very credible downside case.

Exhibit 1

Source: Woodmac, IHS and SSR Analysis

Overview – Too Many Variables

The ethylene industry moved from famine to feast in the last 5 years, but is in a current state of flux so dramatic and unclear that it is very hard to predict how much food will be on the table next month let alone in 4 to 5 years from now when a number of new investments in the US are scheduled to begin production.

The problem is that there are too many variables involved in any attempt to forecast, and this allows for a credible basis for almost any 12 month to 6 year scenario:

  • Crude oil pricing – arguments can be made for a recovery back toward $100 per barrel, but we can also construct an argument that keeps prices where they are today for the long-term. Geopolitics is always a wild card.
  • US natural gas – currently cheap and possibly cheaper if the weather stays relatively mild, but the growth in US natural gas production over the last two years has come from wet gas and gassy crude, with the motivation being the exploitation of the high crude price and the high NGL values (mostly derived from crude). These values are much lower today and no-one can yet afford to frac for dry gas at prices sub $3.00 per MBTU unless they are close to a market that will pay $3.00 per MMBTU. If drilling in Western Marcellus, Eagle Ford and the Permian slow down, and as exports of LNG increase – natural gas prices could reverse.
  • Ethane pricing relative to natural gas – we estimate that US producers of ethylene made an average of 4.0 cents per pound of ethylene just on the discount that ethane had relative to its fuel value in 2014 – this happened because ethane was in surplus – partly production and partly ethylene plant closures, but also because fractionators were making so much money on the other NGL components. This too has changed.
  • Propylene pricing relative to ethylene: the current driver of very low ethylene costs in Asia and Europe is the relative high price of propylene and other co-products. Propylene has traded at a significant premium to ethylene in the US for the last several years and is essentially at parity in Asia and Europe. Propane PDH economics suggest that (if this becomes the price setting mechanism) parity with ethylene is a reasonable assumption even if crude oil recovers.
  • Demand growth: this is the big wild card, though we believe not as big as some would make out or like! We do not believe that the global market is tight enough to warrant pricing in 2015 and 2016 above the cost of the “break-even producer”, which has essentially been the mechanism for the last 4 years. Some derivative markets are so oversupplied globally, such as PVC and polyester, that competition will impact input costs such as ethylene and keep prices at economic lows.

Today is very different; as costs have fallen, pricing, while tumbling also, is lagging considerably. This has opened up very wide margins all around the world and no-one is operating at break-even or even close to it. However, the global markets for ethylene and many of its major derivatives are oversupplied and very competitive and we would expect price declines to catch up with costs as we move through the first quarter of 2015. The first quarter could show some quite good profits but we would not expect to see much beyond that without a step change in demand in 2015 – very unlikely given more and more negative forecasts for global economic growth.

Costs Falling Faster Than Prices – For The Most Part

Ethylene production costs globally have collapsed over the last few months – falling by as much as 60-70% for those producers using naphtha and advantaged propane (propane trading below local fuel equivalent), and still enjoying relatively high prices for ethylene co-products. Interestingly ethylene pricing in the US appears to be falling faster than it is in Asia or Europe, and where we have not seen the same degree of cost collapse for those units running ethane as feedstock, squeezing margins more quickly. US ethylene producers using propane or light naphtha have seen a much more dramatic decline in ethylene costs, because of the value of the co-products and it is possible that ethane may have been the least attractive ethylene feedstock in the US in December! Propane has certainly been the best, with costs in December as low as “negative” 5 cent per pound: in other words the units are being paid to make ethylene.

Outside the US our models show a number of European ethylene units with a variable cost of production close to zero in December, particularly those that can get their hands on low cost propane.

Exhibit 2

Source: Woodmac and SSR Analysis

In the period from September to December 2014, the US producer of ethylene from ethane has, on a relative basis seen a much more disappointing margins trend than almost every other ethylene producer in the world.

However, it is all a matter of timing. Ethylene markets are generally oversupplied, particularly for some of the globally traded derivatives, such as PVC, styrene and polyester, and it will likely be a race to the bottom, with global ethylene pricing falling back into a cost plus relationship with break-even economics. Break-even producers are likely to remain in Europe and Asia, even with the oil and natural gas prices that we see today. US natural gas would need to move up to $6.00 per MMBTU, while Brent remained at current levels for the US ethane based producer to be the global price setter, a position it has held before for brief periods in the early 2000s.

For the last few years average ethylene pricing in the international market has settled roughly $75-100 per ton higher than the theoretical levels predicted by our cost model. This makes sense given that pricing in Asia and Europe are quoted on a delivered basis and our model calculates cost on an ex-works basis – delivery costs for ethylene can be high where there is no pipeline and international prices are generally quoted on a “delivered to somewhere” basis.

For the last several years, ethylene prices in Asia and Europe have been well above those in the US – Exhibit 3. This is part a reflection of the lower costs in the US, but also reflects the net trade position of each region, Asia and Europe are net importers, while the US is a substantial net exporter. The current gap is around 10 cents per pound, but in our models which follow we are assuming that this gap falls to 5 cents per pound, which is a more representative historic US/Europe relationship. If we assume a bigger discount we can get quite bearish about US production economics.

Exhibit 3

Source: Woodmac

Crude Oil – The Big Daddy of Assumptions

While US Natural gas prices, propylene prices, global demand and a host of other variables matter, the possible range of forward oil prices is the most significant variable in determining the fate of the US ethylene industry. As the chart in Exhibit 4 shows, crude oil prices make a huge difference. In the analysis we have kept natural gas prices in the US constant at $3.00 per MMBTU and just varied the price of Brent, allowing Brent to drive international naphtha pricing and ultimately co-product pricing as it drives up the cost of ethylene.

It is likely that the demand line would move under the different scenarios as we would expect slightly more demand at lower prices than higher prices –
see our previous work on ethylene demand elasticity
, but given how flat the curves are at that point, any movement is likely well within the accuracy of the model itself.

Exhibit 4

Source: Woodmac, IHS and SSR Analysis

So – how to think about crude oil? The forward curve on December 31st had Brent moving up to average around $64 per barrel in 2015 and just under $70 per barrel in 2016, but as the Bloomberg screen shot in Exhibit 5 shows, we probably should not rely on the forward curve too much. However, as we are intending to try and model 2015 for US producers we run a scenario in the next section with crude at $64.

Exhibit 5

Source: Bloomberg December 31st 2014

The Bear Case For Crude and The big risk for US Ethylene Investors

We talk a lot about learning curves or experience curves and in prior work at prior firms we created a fairly sophisticated model for the US basic chemicals industry suggesting that

all things being equal, the break-even cost for most large scale products probably declines by between 1-2% each year, as producers optimize and get generally more efficient. However, it is not a smooth curve. The rate at which you look for ways to lower your costs is a function of how much pressure you are under. The rate of improvement for the ethylene industry was far higher in the 1981 to 1985 period than at any other time as margins were low and expected to stay low – Exhibit 6.

Coincidently, this period preceded an equally challenging time for oil producers. Through the early 80s OPEC was holding oil in the low to mid $30s per barrel. This supported lots of projects, especially offshore that had F&D costs in the mid $20s per barrel. When oil collapsed in 1986 the preachers of doom were out in force, predicting the end for most complex offshore drilling. Within a very short time the industry proved them wrong and was back at it, but with much lower costs – costs needed to come down so they did!

There has been just as much talk about a need for $70-80 crude to justify fracking in the Bakken and some of the other US fields as the was about the need for $25 crude to justify drilling in the North Sea in 1985. The current environment will sort out what that real cost is. We have no idea, but we would certainly place a bet on something lower than the range discussed for the last few years. If it is as low as $50, the US new ethylene build economics look very shaky, even with natural gas at $3.00.

Exhibit 6

Source: IHS, Woodmac and SSR Analysis

Natural Gas – It’s all About the NGL values

US E&P companies have not really been drilling for natural gas for the last couple of years, they have, for the most part, been drilling for NGLs and Crude and the gas has come along for the ride. The economics of drilling for the heavier components has been very compelling, with propane, butane, pentane and condensate prices trading relative to crude and at a huge premium to natural gas. The money has been made in the liquids.

Today, US propane is trading below its historic average relationship (33 year average) with US natural gas for the first time since April 2009. Fractionator (the unit(s) which separates ethane, propane etc. from natural gas) margins have all been in the in the propane, butane and other components for almost 2 years, with ethane given way for less than fuel value for the same period – Exhibit 7. Fractionator margins have collapsed as crude oil has fallen and brought the NGL values down with it. While the ethane discounts cannot go much lower than they are already in Exhibit 7, because at some point you would rather burn ethane as a fuel, they can quickly go away and become premiums if drillers slow activity because of lower overall returns. Note that the negative extraction margin average for 2014 contributed 4 cents per pound to US ethylene margins in 2014. Prior to 2013, the 32 year average ethane extraction margin in the US was around 6.5 cent per gallon. If we moved from the 2014 discount to the historic premium in 2015 we would take around 7 cents per pound out of ethane based ethylene margins.

Exhibit 7

Source: Woodmac, IHS and SSR Analysis

Going to the forward curve again (Exhibit 8), we see that natural gas prices are expected to average around $3.25 per MMBTU in 2015 and only around $3.50 in 2016. This may be conservative in a low oil environment as any slowdown in crude and wet gas focused drilling would have a natural gas supply implication longer-term. Given that natural gas prices fell significantly after this chart was drawn, we are probably looking at a 2015 assumption of around $3.10 today.

Exhibit 8

Source: Bloomberg December 31st 2014

Using forward Brent crude for 2015 of $64 per barrel and natural gas at $3.10, and assuming that all ethylene and co-product prices quickly normalize, we get the chart below – Exhibit 9. This throws us a US price average of around 31 cents per pound if we use only a 5 cent reduction from Asia and Europe prices – the price would be lower based on recent discounts. The US ethane based ethylene producer is making around 20-21 cents per pound of cash margin on this basis, versus an average of 36 cents per pound in 2014. Contract based margins were around 30 cents in December 2014.

Exhibit 9

Source: Woodmac, IHS and SSR Analysis

Ethylene Supply/Demand – Growth would Need To Be Extreme Or Production Problems Unprecedented To Change The Conclusions Materially

In our analysis we are making an assumption on ethylene demand for 2015 to create the vertical intersect on the curve. We are using a recent estimate from Wood Mackenzie (published in November 2014), but we think that this likely overstates likely demand, but at the same time we are not sure that it matters too much.

Most recent economic growth forecasts are much more negative than they were 6 months ago for all global regions except the US. For Woodmac’s 2015 demand growth of almost 4% to be correct (Exhibit 10) there would need to be a much greater GDP multiplier in 2015 than 2014. This is possible, given lower crude and expected lower product pricing but it is still a stretch – we would probably place demand growth in the 2.5-3.0% range. In 1987, following an equally dramatic cut in crude oil prices, ethylene demand growth was not much higher than the prior year. Demand jumped in 1988, but this was because of acute shortage, prices increased as inventory built, not because underlying demand was materially stronger.

However, for the purpose of this analysis it does not really matter, as neither projection really matters in determining the break-even cost – given the flatness of the curve at that point.

Exhibit 10

Source: Woodmac

In our analysis we are using demand for 2015 of around 144 million tons of ethylene. This is against capacity of around 162 million tons, suggesting an operating rate of 89% of capacity. In the cost curves we assume 97% operating rates for each facility and consequently total possible production is 157 million tons.

Operating rates of 89% are insufficient to cause market tightness and pricing much above break-even. In the margin peak of 1988, the world was operating at 95% of capacity, but more important there was only 2.5 million tons of surplus capacity in the world (2 world scale units today). In 2015, the world will likely have almost 20 million tons of surplus capacity. As we have discussed many times, a 2015 ethylene peak is a very poor planning assumption.

Estimates and Valuation

We would expect the industry to spend Q1 finding a new equilibrium for ethylene, co-product and derivative prices – all lower – with the rate of decline determined by competitive pressure such as Williams re-entering the US ethylene market and Chinese producers trying to steal export market share for PVC and polyester all over the world. If oil and natural gas prices follow the forward curve – highly unlikely but not sure what else we or anyone else would use right now – we come back to the 15 cents per pound decline in US ethane based ethylene margins that
we first suggested
in work
in October 2014

Since that time, estimates for ethylene producers have generally come down, but not nearly enough – Exhibit 11 – with the exception of Eastman

Exhibit 11

Source: Woodmac and SSR Analysis

Given that EMN has the ability to run its ethylene units on mostly propane, unless there is a collapse in propylene prices in the US – below ethylene price levels (which seems very unlikely until new PDH capacity starts-up), EMN could actually make more money on its US ethylene facilities in 2105 than in 2014. Exhibit 12 shows recent ethylene/propylene price ratios around the world – we expect the propylene ratio to fall in the US as propane cracking increases and this will likely reduce propane based margins and establish a new ethane/propane equilibrium in the US.

Exhibit 12

Source: Woodmac

DOW, LYB and WLK have the ability to run propane in their ethylene units, as do most other US ethylene producers. As a consequence we expect the current advantage for propane to decline, as increased demand in the US plus huge export margins will put a floor under propane, while lower natural gas prices will cut ethane prices further. These lower likely values for ethane and propane add to our concern that we could see drilling curtailed in the US as E&P margins fall.

While 2015 estimates (except for Eastman and probably DuPont) look very vulnerable to lower US ethylene margins, valuations tell different stories – Exhibit 13.

Valuations for DOW, LYB and WLK have come down, with DOW and LYB at what we would call normal value today. WLK is much closer to fair value than it was. If we thought earnings revisions were done and estimates were correct for 2015 we would be more constructive on all three. While the downside is probably limited, without a complete collapse in margins, it is likely that the stocks will go lower before they go higher.

The obvious long idea is EMN; valuation is very low and lower propane pricing helps EMN because it is a consumer of propylene and lower propane pricing means lower propylene costs. Moreover, EMN is taking the propylene downstream into products that are not glob ally traded and in oversupply.

The ethylene buyers, like AXLL and TSE (not included in our valuation analysis because we have no history and no obvious proxy) are all likely to benefit from cheaper ethylene, but for the most part they are in businesses with greater global competition, and this could provide some – and possibly total offset.

If you believe in low natural gas and lower crude the pair trade is long EMN and short WLK; if you believe in low crude and rising natural gas you would still probably be long EMN, but LYB would be as interesting a short as WLK.

Exhibit 13

Source: Capital IQ, SSR Analysis

©2015, SSR LLC, 1055 Washington Blvd, Stamford, CT 06901. All rights reserved. The information contained in this report has been obtained from sources believed to be reliable, and its accuracy and completeness is not guaranteed. No representation or warranty, express or implied, is made as to the fairness, accuracy, completeness or correctness of the information and opinions contained herein.  The views and other information provided are subject to change without notice.  This report is issued without regard to the specific investment objectives, financial situation or particular needs of any specific recipient and is not construed as a solicitation or an offer to buy or sell any securities or related financial instruments. Past performance is not necessarily a guide to future results.

Print Friendly, PDF & Email