The rapidly declining production of small Dutch gas fields

The rate at which gas production from Dutch small fields declines is much faster than the rate at which Dutch gas consumption declines. Dutch gas is effectively being replaced by Russian gas. This is not in the best interest of the Netherlands, neither from a financial nor from an environmental point of view. Current policies will not be able to preclude a rapid and near complete collapse of gas production from Dutch small fields.

Introduction.  Dutch gas production has two components: the production from the giant Groningen field and the production from numerous small fields. The rapid decline of production from the Groningen field, due to production measures put in place to limit seismicity, has received a lot of attention. That the production from small fields is also rapidly declining, for very different reasons, has hardly received any attention.

For decades the Dutch government has stimulated the production from small Dutch gas fields. The aim of this successful “small fields policy” was to maximise revenue from Dutch gas and at the same time preserve the Groningen field as much as possible.

Prior to the year 2000 the production from small fields (both offshore and onshore) exceeded 40 BcM per year. After 2000, a gradual and slow decline started to set in, which was compensated by increasing the production from Groningen. By 2007, production had decreased to about 35 BcM and fell below that of Groningen for the first time in decades. By 2012, production had decreased to about 30 BcM. Until this time the decrease was solely due to geology. The early (larger) finds started to deplete and later finds gradually decreased in size.

A turning point for Dutch gas production: 2012.  In recent years there has been a marked change in the operating environment for Dutch gas production. Prior to this shift Dutch gas was seen as a welcome source of revenue for the Dutch government, obtained from the production of a relatively clean fossil fuel. After this shift, gas became a polluting fossil fuel which one would only like to tolerate for a limited time, awaiting the completion (sooner rather than later) of the energy transition.

A number of elements play a role in this shift:

  • The increasing realisation of the severity of the climate change problem and the increasing momentum to actually start tackling this problem, culminating in the Paris COP21 agreements.
  • The increasing magnitude of Groningen earthquakes and the plight of people affected by these earthquakes, culminating in the 2012 Huizinge earthquake that damaged thousands of houses. This damaged the public image of gas in general and the image of the largest producer (NAM, a Shell ExxonMobil joint venture) in particular.
  • The increasing unpopularity of large corporations such as oil and gas companies, perceived to make profits at the expense of local populations.

All this has been a gradual development. If I would need to pick a turning point though I would place it in 2012.

The figure below shows the production from small Dutch gas fields until 2012 and the potential scenarios for future production at this point in time. A large range of scenarios was possible (depending on future gas prices, the amount of government support for small field production and the success of several exploration plays). What has so far materialised is a scenario with a very low production from small fields. This decline is more severe than in the past and is no longer just related to geology.

small-fields

Recent developments.  The last few years have seen only small additions from new fields (with 2016 being an absolute low point). Exploration for new fields is rapidly declining. It seems increasingly likely that a number of operators will cease to explore altogether. In addition to geological factors (long term creaming of the area) a number of additional elements come into play, creating a perfect storm for Dutch small field gas production:

  • Low gas prices
  • The absence of support from the Dutch government
  • Doubts on the long term future of the Dutch offshore gas infrastructure system
  • Obtaining a permit for onshore drilling has become a very tedious and time consuming procedure; obtaining a permit for a new onshore production location has become even more difficult.

For the offshore gas production this can have a snowball effect. If an increasingly smaller number of fields has to carry the operating cost of an entire offshore pipeline system, at some stage the moment will arrive that this is no longer commercially feasible – increasing the amount of gas that is left in the ground. This moment is now rapidly approaching, especially in the case of continuing  low gas prices.

Consequences for the Netherlands.  The difference between the current and the late 2012 mid-case production profiles is about 170 BcM. Depending on future gas prices this represents a value of some 15 to 30 billion € (primarily to the Dutch state). I would estimate that by now roughly half of this volume has been irrevocably lost; half could still be saved if adequate measures are now taken in a world where oil and gas prices are slowly recovering. The public discussion on this issue has been minimal.

Whilst the energy transition will be a matter of decades, with an expected gradual decline of gas consumption in the Netherlands between now and 2050, the current decline in gas production in the Netherlands has become much more dramatic. Dutch gas is now being replaced by Russian gas and/or coal. That is not in the best interest of our country, neither from a financial nor an environmental point of view. Short term targets on emissions are becoming more difficult to achieve.

Coal is the most polluting fossil fuel; something that can only be mitigated, to a limited extent, by costly additional measures. Russian gas implies both additional CO2 emissions (roughly 12 % of the gas is needed to transport gas from Russia to The Netherlands due to the low efficiency of the Russian gas transport system) and additional methane emissions (methane losses related to transport over large distances in Russia are expected to be significantly higher than methane losses related to transport over short distances in The Netherlands). Effective emissions (CO2 equivalent) are estimated to be about 20 – 25 % higher compared to Dutch gas.

In conclusion, I would advocate measures that would stimulate the production of remaining gas reserves such as a preferential tax treatment for small fields or fields with low quality reservoir. The Dutch tax regime for gas producers is significantly worse than for instance the tax regime in the UK (50 % versus 30 % direct tax take). The policies that are currently in place are grossly inadequate to preclude a rapid and near complete collapse of gas production from small fields. Part of the revenues could be used to fund a much needed, but also costly and lengthy, Dutch energy transition.

Posted in Geen categorie | Leave a comment

Oil markets: a turbulent 2016, an uncertain 2017

After two and a half years of opening up the taps (or rather: not closing them) OPEC has changed course in what is looking to be a gamechanger for the oil market. Market sentiment has shifted and the oil price has gone up by some 20 %. We can look back at a turbulent 2016 and look forward to an uncertain 2017.

OPEC defeated?

Some say OPEC’s decision to cut was a defeat. Was it really? OPEC (and most of all: Saudi Arabia) over the last two years has been trying to deal as good as possible with the difficult situation that the 2009-2014 high oil price world had created for them.

Would they, from 2014 onwards, have defended price instead of market share US tight oil production would have risen by about 2 mb/d by now (instead of the reduction of about 1 mb/d that actually materialised). The decline from non-OPEC conventional fields would have been 3 mb/d (instead of the 6 mb/d that materialised). For Saudi Arabia it would have been a repeat of the early 1980’s when they did defend price, resulting in a reduction of their production to a level as low as 2.5 mb/d before they gave up.

Two years of defending market share instead of price has resulted in a non-OPEC supply 6 mb/d lower compared to what it would have been otherwise. The large investment cuts in non-OPEC oil will reduce non-OPEC supply for years to come. That is major progress for OPEC. It has brought supply and demand close to equilibrium in 2017. Now a cut became a realistic option in order to bring higher oil prices forward. A 2014 cut would only have postponed the inevitable and increased the length of the subsequent painfull rebalancing period.

They will be disappointed by the resilience of US tight oil. US tight oil has been saved by drilling in the very best spots only, increased efficiencies and reduced service industry costs. Furthermore it has been saved by their investors and financiers – for whom accepting severe losses was a better alternative than to let them go bankrupt and cease operations.

OPEC has regained market share and, more importantly, some of their ability to move markets. US tight oil has survived with break even costs in the very best areas that are now at the lower end of the global non-OPEC cost curve. They have both paid a heavy price. But it is high cost non-OPEC conventional oil that has lost the most in this battle.

Why cut now?

First and foremost, markets have done their work and supply and demand have been approaching a balance, enabling a meaningful cut.

Budget deficits have troubled all producers. A country like Venezuela has been desperate for a deal. Unfortunately for Venezuela it has no clout whatsoever in OPEC. The defining push for the agreement has been given by Mohammed bin Salman (Saudi Arabia’s de facto ruler) and Vladimir Putin. The situation within their countries is such that both have good reasons to do so. MbS is aiming to solidify his grip on power. For that he needs to limit hardship for the Saudi middle class and provide hope for the rapidly growing (and increasingly unemployed) number of young Saudis. Saudi Aramco’s planned IPO will benefit from higher oil prices. Putin as well wants to limit hardship for the Russian population. He can not be as indifferent to the wellbeing of the Russian population as Stalin once was; his grip on power is more secure if he keeps the Russian middle class happy.

Saudi Arabia needed to see pledges from other producers (Iran and Russia in particular) to go ahead. These other producers needed to have confidence that limited cuts will give a substantial increase of the oil price – something that the initial market reaction in September upon the Algiers talks provided to them. Reaching a Vienna agreement became a must; not reaching it would have implied a substantial price drop, something they could ill afford.

OPEC spare capacity is at its lowest level since 2008. Iran is back at its pre sanction level of production and cannot raise production any further in the short term. Russian production is at a record level. Even if producers do not fully live up to their pledges, their ability to cheat and take away market share from Saudi Arabia has become limited. Saudi Arabia can be satisfied that cuts are shared. Iran and Russia can be satisfied as well; their pledges are not a great hardship for them.

What has changed?

Saudi Arabia has lost clout within OPEC. Iraq and especially Iran are challenging its dominant position. Their combined production starts to approach that of Saudi Arabia. Both have large undeveloped oil reserves, which can be developed at low cost, and are still producing way below their potential. In the long run they are likely to further ramp up production. After having been sidelined for a long time due to wars and sanctions both are now reclaiming their natural position in the OPEC pecking order. In the long term, reaching an agreement within OPEC will not become any easier.

Iran in particular by now seems in a better position to overcome periods of low oil prices than Saudi Arabia. Its economy, hardened by years of sanctions, is better equipped to do so and is less reliant on oil income. Saudi Arabia’s pivotal role in OPEC was based on its being the largest producer by far, its maintaining a substantial spare capacity and its large financial reserves that (in combination with a relatively small population) enabled it to better sit out a prolonged period of low oil prices. Some elements of this dominance are now starting to fall away and Saudi Arabia is no longer the sole pivotal nation within OPEC that it used to be.

Oil has always been linked to politics. Saudi Arabia has lost political clout in the Middle East. They are struggling to deal with a Shia encirclement. Their economy is solely dependent on oil and and is not performing well in comparison with countries like Dubai. The religious establishment is a blocker when it comes to reforming education and increasing the role of women in the economy. Their special relationship with the USA is deteriorating now that the USA is moving towards energy independence and more reluctant to prop up fundamentalist regimes. In the meantime Russia has gained influence, by intervening in Syria and by playing a key role in brokering the recent OPEC agreement.

For a long time Saudi Arabia has been a source of stability in the region and Iran a source of instability. That is changing now.

Where do oil prices go from here?

All ingredients seem to be in place to keep oil prices in 2017 at a systematically higher level than in 2016. Compliance will vary among the different producers. But the over 700,000 b/d cut from Saudi Arabia and other members of the GCC (Kuwait and the UAE) in itself is sufficient to bring supply and demand approximately into balance. Even if the level of compliance of other producers is as low as 50 %, the cuts will still lead to a meaningfull reduction of oil inventories.

As oil producers see that the deal delivers the promised rise in oil prices, they will be less likely to cheat – at least initially. The pledged, gradual cut from Russia involves little more than natural decline and a reduction or freeze of short term projects. Russian oil production has been maximised as much as possible over the last few months in 2016, to a level that may be difficult to sustain anyway in the first half of 2017.

Volatility will be there to stay as stories of cheating, outages and potential production increases from uncapped Nigeria and Libya abound. Notes to refiners about shipment reductions will be duly leaked to the media (something that has already started). But throughout the year we will be seeing a return to oil prices that are closer to the long term sustainable price of oil: that of the marginal non-OPEC barrel, somewhere near $ 60 – 80 per barrel.

And let us put things into perspective. The oversupply frequently described as a glut was no larger than about 2 % of global production, at its worst. OECD oil inventories have hovered around 65 days of supply in 2016; little more than 5 days above the long term average. It does not take that much oversupply to send oil prices plummeting. In the same way, it does not take that much undersupply to send them through the roof. One should not blame analysts too much for not being able to predict the oil price. But a bit of blame for underestimating uncertainties might be justified.

How will US tight oil react?

A main source of uncertainty is how US tight oil will react to higher prices. Tight oil’s shorter cycle time and a faster reaction to changes in oil price compared to conventional oil may keep a lid on oil prices. But to what extent?

If we compare a recent global cost curve of oil projects with a 2014 cost curve there are two developments that stand out. Firstly, the average breakeven cost has decreased substantially, from about $70 per barrel in 2014 to about $50 per barrel today. Secondly, over the last two years US tight oil has seen the biggest cost decreases and it has shifted towards the left (more competitive) side of the global cost curve.

For those US tight oil companies that have survived and that have quality acreage there now seems to be a great promise for the future: break even prices near the lower end of the global spectrum of opportunities and huge in place volumes. This is the background for the recent outperformance of share prices for companies active in the Permian (the US region with the lowest break even prices). Break even prices for the very best areas have dropped to about $30 – $40 per barrel. As a whole the US tight oil industry is estimated to need about $55 – $60 per barrel to maintain a flat production level.

There is one snag: break even prices quoted above are for current cost levels of the service industry, widely seen as being unsustainably low. How much will these costs increase once that activities pick up in earnest? Rystad Energy estimated that for the Bakken about 40 % of cost savings were structural (faster drilling, better well production) and about 60 % of cost savings were cyclical (primarily lower service industry costs, to a lesser extent drilling in the very best sweet spots only). When activities pick up significantly, break even costs are expected to increase by about 65 %. Current break even costs for the very best sweet spot areas would be expected to increase from $ 30 to $ 50 per barrel. Non core areas (that currently see little activity) could see an increase from $ 50 to $ 75 per barrel. Other studies have reached similar conclusions.

Quoting from a recent SPE panel discussion: “If oil prices stay below USD 55/bbl, equipment availability can be relatively smoothly managed in the Permian. But at prices from USD 60/bbl to 70/bbl all of a sudden all of the other plays come back, and then for sure we reach the threshold of equipment not being available”.

I feel that many analysts overestimate the ability of US tight oil to act as a swing producer. Firstly, things take time. Hiring drilling crews to man the often less efficient rigs that have now been cold stacked takes time. Hiring fraccing crews takes time. Getting permits takes time. It took two years before the effect of low oil prices on US tight oil production had materialised in full. Secondly, costs of drilling and fraccing follow the oil price. US tight oil will indeed keep a lid on oil prices. But to a smaller extent than what is often assumed.

Posted in Geen categorie | Leave a comment

Groningen gas production and earthquakes

 

  1. Introduction

The giant Groningen field was found in 1959. Remaining reserves are about 800 BcM (vs. about 2800 BcM of initial reserves). The field is exploited by NAM (50 % Shell, 50 % ExxonMobil) on behalf of the “Maatschap Groningen”, a partnership of NAM (60 %) and EBN (40 %). EBN is wholly owned by the Dutch state.

Including the effect of taxes, 85 – 90 % of the profits from the field go to the Dutch state. Income from gas production (which is for about 80 % coming from Groningen) accounted for about 10 % of state income over the 1975 – 2015 period.

Over the years earthquakes related to gas production have increased in number and magnitude. The shallow depth of these earthquakes (as well as the frequent occurrence of a very low velocity layer just below the surface) implies that earthquake intensity and damage to houses are relatively large compared to the seismic magnitude. Older houses and farms built with single brick walls are especially vulnerable. A turning point was the 2012 Huizinge 3.6 magnitude earthquake which damaged thousands of houses. This was followed by the realization that the estimate for the maximum potential magnitude of 3.9 for a Groningen earthquake, assumed so far, was likely to be too low (and subject to considerable uncertainty) and that seismicity could not only result in material damage but potentially also in loss of life.

gron-earthquakes-numberNumber of earthquakes exceeding a certain threshold magnitude as a function of time (from Muntendam and de Waal, 2013). It is likely that light tremors already took place during the late 1980’s (the seismometer network has been gradually extended and is only expected to have detected all tremors with a magnitude greater than 1.5 since the mid 1990’s).

As a result the Dutch government has put in place a production cap for the field (as well as separate caps for the area’s most affected by seismicity), leading to a significant decrease in production and government income. By October 1, 2016, the production cap has been lowered to 24 BcM per year (in 2013, the last year prior to the recent production caps, yearly production was still as high as 54 BcM). The double hit from production caps and declining gas prices has resulted in a severe decrease of government income from about 15 bn Euro (2013) to little over 2 bn Euro (2016).

aardgasbaten

  1. Groningen earthquakes through time

The first registered earthquake in the Northern Netherlands related to gas production took place near Assen in 1986. The first registered earthquake from the Groningen field occurred in 1991. In 1993 a joint NAM, KNMI (the Dutch meteorological institute – which also has a section monitoring seismicity in the Netherlands) and SODM (the government entity supervising the oil and gas industry) study (in Dutch) confirmed the relation between gas production and earthquakes. Prior to 1986 no earthquakes whatsoever had been recorded in the tectonically quiet Northern Netherlands.

What has not helped the public image of the NAM in the long term is that from 1986 to 1993 they denied a relation between gas production and earthquakes. Instead of admitting that when earthquakes start to appear at producing gas fields (in an area without any recorded seismicity so far) there is very likely to be a relation (be it that the way this works is not well understood) they went into denial mode.

This denial mode continued for many years. The 1993 joint report contained a section in which an estimate was made of the maximum potential magnitude of a Groningen earthquake. This method (using a Gutenberg Richter relation between earthquake magnitude and frequency) played a key role in KNMI reports with estimates of maximum magnitudes for years to come. The problem is that deriving a maximum potential magnitude from historically observed seismicity, using the Gutenberg Richter relation, is something that can only be done for stationary situations (as is usually the case for natural earthquakes). A depleting gas field, however, is not a stationary situation at all. If anything it gave a lower bound for the maximum magnitude of a future Groningen earthquake. For Groningen, as depletion progressed, the KNMI gradually revised its estimates from an initial 3.3 (1993) to an eventual 3.9 (2006), upon which it eventually became increasingly clear that this method was not defendable.

For a long time both NAM and SODM were not proactive in starting further research (for instance to determine uncertainty estimates on KNMI’s estimates or to invite alternative views from other research organizations). The focus of the NAM was very much on subsidence rather than on seismicity. An in depth review is given in a report of the Dutch Safety Board on Groningen earthquake risks. There was a genuine expectation in NAM that earthquakes could only result in limited material damage. This was not backed up by solid research, however, and in hindsight uncertainties were severely underestimated.

 

  1. Understanding Groningen earthquakes: differential compaction related to faults

The Groningen gas is contained in Rotliegend sandstone reservoir (in the pores in between the sandstone grains). The gas originates from deeper Carboniferous coals. As these coals experienced increasing pressure and temperature (while gradually being buried deeper over geologic time) they started to expel gas which moved upwards (due to its low density) to the overlying sandstones. The sandstones in turn are overlain by Zechstein salt which is impermeable. As a result gas accumulations exist in those places where the Top of the Rotliegend sandstones is at a relatively shallow depth (i.e. more shallow compared to surrounding areas) so that the gas, which cannot move upward through the salt, cannot escape laterally.

groningen-3d-view3D view (from the SW) of the Top Rotliegend. Colors denote depth (red is highest); blue plane denotes the Gas Water Contact; red lines denote wells. Source: NAM MMax workshop.
groningen-xsE-W cross sections across the Groningen field, illustrating how faults offset the Rotliegend reservoir layer (in yellow). Red layer denotes the Ten Boer shales above the Rotliegend. Zechstein salt immediately above the Ten Boer; Carboniferous below the Rotliegend sands. Source: NAM MMax workshop.

As gas is being produced, the pressure in the gas decreases and the overburden of about 3 km thickness starts to exert a greater pressure on the sandstone grains. As a result, the sandstone starts to compact. For the Groningen field the total amount of reservoir compaction due to gas production is expected to range up to about 50 cm (for a layer that is up to about 250 m thick). If compaction were to take place homogeneously over the field significant seismicity would be unlikely to arise and the main adverse effect of gas production would be the gentle subsidence bowl that we see developing in the Groningen area since the 1960’s. Unfortunately it does not. The Groningen field is affected by numerous faults and these faults can lead to strong local variations in compaction. As differential compaction over a fault increases, shear stresses on the fault plane build up and at some stage the strength of material at the fault (a zone of weakness) is overcome after which rupture takes place on (part of) the fault plane.

There are a number of ways in which faults can lead to differential compaction. The most straightforward one is that faults offset the gas bearing reservoir vertically. As a result compaction at a given depth varies laterally (shown schematically in the figure below).

groningen-fig-1-schem-extraSchematic representation on how earthquakes can be triggered in a compacting, faulted reservoir. The actual situation for Groningen is much more complicated. The thickness of the Rotliegend reservoir sands varies considerably over the field (from about 80 to 250 m) and over the field area some 1600 faults have been mapped (a few large ones and many small ones).

In addition faults can be baffles (if not hard boundaries) to flow. As a result pressure differences across faults may arise which can result in differential compaction and seismicity. Reservoir simulation models of the Groningen field point to some faults that are indeed sealing. Regionally, the NW-SE trending strike slip faults are often observed to be sealing faults. Until recently NAM’s production philosophy has been to minimize pressure differences across the field (helped by the excellent permeability of the Rotliegend reservoir). As a result it seems unlikely that pressure differences between major compartments make a large contribution to differential compaction and Groningen seismicity.

Finally, faults should not be seen as sharp boundaries. A larger fault tends to be surrounded by many smaller ones as well as a zone (“damage zone“) where reservoir quality is impaired and porosities are substantially reduced. For faults in similar, smaller Rotliegend fields in the Dutch offshore such damage zones are observed to be about 50-100 m wide, with an average porosity reduction of about 3 p.u. (ranging up to about 5 p.u.). With compaction being strongly, non-linearly, dependent on porosity such a reduction in porosity can easily result in a reduction of compaction by roughly a factor 2. Damage zones may be the most pronounced for strike slip zones which have experienced major displacements (but do not necessarily stand out on seismic as their vertical displacement may be limited).

 

  1. Key observations on Groningen seismicity

Significant progress on our understanding of Groningen earthquakes has been made over the last few years. The network of seismometers has been extended and now includes seismometers at depth, reducing the uncertainty of earthquake locations. We can now be more confident that earthquakes are taking place at (or in the immediate vicinity of) the Rotliegend reservoir. Two key empirical observations are listed below.

Seismicity (initially) increases with total reservoir compaction. A threshold exists of about 10 – 15 cm of reservoir compaction before the onset of seismicity. Beyond this threshold seismicity starts to increase in both number and magnitude.

Apparently this is the amount of compaction needed to bring shear stresses on faults (or at least on some locations, on some faults) to the critical level required for rupture. Initially, stresses on faults (at least at Rotliegend level; likely also at e.g. deeper Carboniferous levels) are not anywhere near the level required for rupture. This is in accordance with the complete absence of natural earthquakes in the Northern Netherlands.

 groningen-time-series-b-en-oTime series of M > 1.5 earthquake magnitudes versus reservoir compaction at the origin time and epicentre of each event (from Bourne and Oates, 2014).

What is less clear is how (if at all) this trend will continue in the future. As long as most earthquakes are related to the first slip event on a fault (segment) seismicity is indeed expected to continue to increase with the total amount of compaction. But once most earthquakes become related to a second or later slip event on a fault (segment) the increase of seismicity with compaction will start to diminish and we may even start to approach a stationary situation.

The figure below shows the fraction of energy related to compaction that has been released seismically (“partitioning fraction”) as a function of the total amount of compaction. Although far from being conclusive (the highest amount of compaction has been reached in a relatively small area only) it cannot be excluded that the partitioning factor is levelling off at a total compaction greater than 20-25 cm. The absence of earthquakes during the first 20 years of production turned out to be no guarantee that earthquakes would remain absent over the entire field life. In the same way, the increasing intensity of seismicity as observed over the 1991-2012 period is no guarantee for its continued increase in future. Our predictions on future seismicity are to a disappointing extent still based on statistics and extrapolation of known trends (rather than a complete understanding and geomechanical modelling of the actual physical processes).

groningen-fig-partitioning-2Partitioning fraction (seismic moment divided by total reservoir compaction moment) for different compaction classes. Green line (NAM) and dashed blue line (TNO) denote relations between the partitioning fraction and reservoir compaction for different models. Note the large amount of uncertainty for the future partitioning fraction in the forthcoming late stage of the Groningen field life. Source: TNO report R11953 (December 2013, in Dutch). It is quite conceivable that the partitioning fraction continues to increase (at a similar rate as observed so far). It is also quite conceivable that the partitioning fraction is levelling off at a value of about 10-4.

Seismicity increases with fault intensity. Within the area of high compaction at reservoir level, seismicity is the most pronounced in areas with a high intensity of faulting. The intersection of numerous NW-SE trending faults with the high compaction area near Loppersum experiences the highest seismic intensity. In comparison, an area with a similar high amount of compaction and a low fault intensity further towards the N experiences a much lower level of seismicity. There is no obvious relation between seismicity and fault throw.

static-model-mapCompaction (denoted by color; color scale ranging from 0 to 0.4 m). Black lines denote faults from the NAM static model. Red line denotes outline of the field. Seismicity is denoted by symbols (size of the symbol indicates event magnitude). Location uncertainty of seismic events is about 1 km.  Highest seismic activity takes place in a region of high compaction and high fault intensity. Source: TNO report R10755 (based on the NAM 2013 static model).

These observations (and in particular the observation that there is no clear correlation between fault throw and seismicity) indicate that fault damage zones may play an important role in Groningen seismicity. The Groningen field is intersected by zones of anastomosing faults, indicative of strike slip faulting (in particular in the Loppersum area and in the Eemskanaal area). Elongated damage zones in these areas (with a much lower amount of compaction) can account for the observation that seismicity is distributed over a larger area (rather than being concentrated on the few large throw boundary faults) and is associated with faults that often do not have a large vertical displacement.

Wells in these areas do sometimes intersect such a damage zone (the clearest example being the EMK-2 well). When incorporated in a static model such wells may erroneously influence the reservoir properties in the model over a large area (resulting in a pronounced mismatch between modelled and observed subsidence).

The observation that there is no clear correlation between fault throw and seismicity critically depends on the correctness of KNMI’s estimate of earthquake location uncertainty of about 1 km. If this uncertainty is severely under estimated it could be that events in reality do primarily take place on the large throw faults. Future work, based on the recently extended network of seismometers, should enable us to better delineate Groningen earthquake locations.

 

  1. Time dependent processes such as creep play an important role.

There is a clear correlation between production and seismicity on a seasonal basis. There is a certain time lag (of about 3 – 6 months) in between, however, and time dependent processes such as creep play an important role.

groningen-seasonalityMonthly Groningen gas production (gray dashed line) and smoothed earthquake event rates (black solid line). Event rates clearly correlate, with some time-delay, with the seasonal pattern in production rates. From Bierman et al., 2015.

Already during the early phases of production it was recognized that subsidence initially took place at a much lower rate than expected (based on laboratory compaction experiments). This effect was only properly understood upon Hans de Waal’s work at Shell’s research lab (and following thesis at Delft University) on rate dependent compaction of sandstone reservoirs. To date, this work is still the basis for most of the Groningen subsidence models.

As a result of time dependent deformation of the Rotliegend it is not clear to what extent production measures will result in an additional reduction of seismicity. In other words: will a reduction in production rate by a factor 2  result in an end member reduction of seismicity by a factor 2 only, per fixed time interval, or (and to what extent) will there be an additional reduction as creep reduces the partitioning factor?

 

  1. Uncertainties are there to stay

In spite of all the advances in recent years, both regarding observations and modeling, our understanding of earthquakes related to gas production – and in particular our ability to predict future earthquake intensity – remains limited.

Maximum potential seismic magnitude.  A workshop on the maximum potential magnitude of Groningen earthquakes was held in March 2016. The outcome from this workshop is a range that is larger than ever before; spanning from 3.8 to 7.25. Not all values within this range are equally probable though and the key question is whether fault rupture will be (mostly or completely) contained within the Rotliegend reservoir or whether rupture can take place over much larger surfaces in the underlying Carboniferous (whether tectonic or triggered by Rotliegend earthquakes).

Should earthquakes remain (largely) confined to the Rotliegend (which is likely – but exact how likely is something for which estimates differ) then the range for the maximum potential magnitude is estimated to be about 3.8 – 5.0. The dimensions of the Groningen field, the thickness of the Rotliegend and the maximum expected pressure drop imply that in this case a Groningen earthquake is unlikely to exceed a magnitude of 5.0.

groningen-mmax-logic-treeExample of a logic tree for the maximum potential magnitude of a Groningen earthquake, as presented at the MMax workshop (from the ExxonMobil contribution). For different contributions the exact ranges and probabilities may differ. The final range of 3.8 – 7.25 represents the full range of all individual contributions. Personally I would put in a much lower probability for Groningen earthquakes to propagate significantly into the Carboniferous.

Effect of recent production measures.  The production caps that have been put in place contain 2 different elements: an overall production cap and specific production caps for the Loppersum and Eemskanaal areas.

For the overall production cap the key question is whether such a production rate decrease will merely result in the same amount of seismic energy now being released over a longer period (which is what the models in the NAM winningsplan tend to predict) or will also result in a reduction of the total amount of seismic energy to be released (which is what is suggested in the SODM reaction to the NAM winningsplan).

For an area production cap such as the one for the Loppersum area the key question is whether this only buys this area a temporary reprieve (and seismicity resumes once that the pressure decline resumes – something that will happen within about a year given the good overall connectivity in the area) or whether a more gradual pressure decline in the Loppersum area will result in a long term reduction of seismicity as well.

TNO expects that for a more gradual pressure drop (primarily due to a smaller seasonal variation in production; for the Loppersum area also due to production taking place at a larger distance) seismicity will be reduced. As yet this is an expectation – which may or may not be confirmed by observations.

 

  1. How to deal with Groningen earthquakes?

Regardless of the eventual outcome I would argue that the uncertainties mentioned above can be managed:

– seismic intensity and maximum observed magnitudes have been observed to increase gradually. Although it cannot be completely excluded an event with a magnitude much greater than what has so far been observed (e.g. including a significant slip component in the Carboniferous) seems quite unlikely.

– production measures work. Local production caps in a high risk region have a marked effect on seismicity within months.

This should enable us to manage production with a hand on the tap like strategy (like for the Waddenzee production) in a responsible way. Currently, the risk of a fatality is estimated to be very small (<< 1). Should risk levels stay roughly at this level the number of fatalities over the coming 30 years would be expected to be of an order of magnitude of 1. For comparison: the number of traffic deaths in Groningen over the coming 30 years is expected to be of an order of magnitude of 1000. Large uncertainties exist for these estimates. But at least we have a calibration point: the last 5 years (with a risk likely to have been greater than the risk over the coming years) did not result in any injuries or fatalities.

 

  1. Alternatives are more costly, less environmentally friendly

A study by CE Delft (an engineering consultancy) looked into the consequences of a number of alternatives for Groningen gas for the environment and for Dutch state income. Alternative sources considered were Russian gas (capacity wise much more feasible than Norwegian gas), LNG (from Qatar) and gas from yet to be developed small Dutch offshore fields. In addition the effects of a reduction in gas consumption were studied.

These alternatives imply a substantial loss of income to the Dutch state. In addition, with the exception of a reduction in gas consumption, they are also less environmentally friendly.

Russian gas, for instance, implies both additional CO2 emissions (roughly 12 % of the gas is needed to transport gas from Russia to The Netherlands due to the low efficiency of the Russian gas transport system) and additional methane emissions (methane losses related to transport over large distances in Russia are expected to be significantly higher than methane losses related to transport over short distances in The Netherlands). Effective emissions (CO2 equivalent) are estimated to be about 25 % higher compared to Groningen gas.

From an environmental point of view a reduction in gas consumption is by far the preferred option. It is not going to happen in the short term, unfortunately. Groningen gas production has already been cut by approximately 50 %; the energy transition will take decades.

groningen-figs-2-ce-alternativesConsequences for emissions and Dutch state income of a reduction of Groningen gas production by 10 BcM per year. Alternatives for Groningen gas considered are Russian gas, LNG (Qatar) and new Dutch offshore gas fields (questionable whether they can have a substantial impact – given the limited exploration successes in recent years in the very mature Southern North Sea). In addition the effects of a similar reduction in gas consumption are studied.

Another option is to reduce seismicity by maintaining reservoir pressures at a higher level. The most straightforward method to reduce the reservoir pressure drop is by nitrogen injection. This has been studied extensively by NAM in recent years and has been rejected as:

  • This is very high cost
  • Involves a large scale industrial project that will take at least until the mid 2020’s before first injection
  • Involves major CO2 emissions and industrial activities throughout the region
  • Will result in a reduction of gas recovery, with some hydrocarbon gas being bypassed by nitrogen.
  • May well result in adverse effects; large scale injection may also lead to seismicity and it cannot be excluded that no net reduction of seismicity is reached.

 

  1. The number of damage claims is rapidly increasing due to people putting in claims for damage unrelated to earthquakes.

Over the past few years the number of damage claims has rapidly increased to over 500 claims per week on average. This increase is not related to an increase of seismicity. On the contrary, the total seismic energy released (per year) peaked in 2012 and has substantially decreased in the following years (see figure below). The largest earthquake in 2012 was the 3.6 magnitude Huizinge earthquake; the largest 2015 earthquake was a much smaller 3.1 magnitude event near Hellum.

groningen-figs-3a-damage-claimsa) Cumulative number of claims as a function of time. b) Average number of claims per week as a function of seismic energy released (both per year) for the 2012 – 2015 period. From the 2016 Technical Addendum to the Groningen Winningsplan.

It can thus be inferred that either a lot of actual damage in 2012 did not result in a damage claim or that a lot of 2015 claims were not related to damage caused by earthquakes. Additional data indicate that the latter explanation is by far the most likely. The figure below shows the percentage of buildings with claims plotted against PGA (Peak Ground Acceleration) for the 2012 Huizinge and 2015 Hellum events. It is expected that houses close to the epicenter and subjected to higher PGA’s have a higher chance of being damaged. For the Huizinge earthquake there is indeed a strong correlation between PGA (which in turn has a strong relation to distance to epicenter) and the percentage of buildings with damage claims. For the Hellum earthquake on the other hand no such correlation is observed and the vast majority of claims come from areas at a large distance from the epicenter with minimal PGA.

groningen-figs-3b-huizinge-hellum-newPercentage of buildings with claims plotted against PGA (Peak Ground Acceleration) for the 2012 Huizinge and 2015 Hellum events. From the 2016 Technical Addendum to the Groningen Winningsplan.

In simple words: in 2012 people put in a claim when a crack in their house appeared after an earthquake that had been felt in the area. The large increase in claims in the following years is a reflection of the intense publicity, the ease with which claims can now be submitted and the calls from NGO’s and politicians to put in claims rather than an increase in actual earthquake related damage. That situation is now becoming difficult to manage. The cumulative number of claims of over 60,000 implies a huge effort, at a significant cost to society (the cost of evaluating claims by now outweighs the cost of strengthening houses and compensating damages). The rapidly increasing share of rejected claims add to the disappointment and disillusionment that many people in the area already experience.

 

  1. The Dutch government now needs to rise to the occasion. 1) Make choices on the basis of a cost benefit analysis. 2) Accept that a situation that all the benefits are for the country, all the downsides are for the local gas producing region is not fair

 

To govern is to make choices. The Dutch government should have the courage to compare the risk that people in the Groningen area are running due to earthquakes with the risks that other people in The Netherlands are running. Flooding is a risk that many people in The Netherlands are subjected to. Risk levels for flooding are of similar magnitude as those for Groningen earthquakes. And yet we make a conscious decision not to raise our dikes to a level that gives absolute security. The reason is simple: cost. Not every medical treatment that is possible is being given – even it would extend or save lives. Again the reason is simple: cost. However difficult to accept this may be: with the limited means that we have there is simply no other choice.

I would argue that decisions on Groningen production caps need to be made on the basis of a cost benefit analysis – as is common practice for other (industrial) activities that involve risk. The cost for the Dutch state of billions of euro’s (even if the gas not produced now would eventually be produced in 20 or 30 years from now) by now seems to become disproportional to the risk to human life that these earthquakes pose (for comparison: a single investment of 120 million euro in Groningen provincial roads would be expected to save approximately 5 lives on a yearly basis). The Dutch Council for Security rightly condemned that safety did not play a role in decision making on Groningen production until only a few years ago. But by now there is a risk that we are going to the other end of the spectrum: that safety with respect to Groningen earthquakes needs to be achieved at all cost.

I would argue that the current unrest in Groningen is not just related to earthquakes but also to a long standing feeling that they are being badly treated by the central government. And here they have a point. Peripheral areas in The Netherlands, such as Groningen, tend to be under-represented in the Dutch parliament. When investments from FES, the Dutch infrastructure fund which received 40 % of Groningen gas income for the 1995-2009 period were analyzed it was found that close to 90 % went to the Randstad area (Amsterdam – The Hague – Rotterdam); only about 1 % of the investments went to the three Northern provinces combined.

In general there are good reasons why the proceeds from mineral wealth should go to a country as a whole. But for this specific case, where a single field accounts for close to 10 % of state income for a period of 40 years and where the downside to the local population is long standing and substantial, this just does not seem fair to me. It seems justified to use a part of the Groningen revenue to establish a fund that solely supports the Groningen local economy and infrastructure.

 

Posted in Geen categorie | 1 Comment

Chinese oil companies: giants on shaky foundations

Abstract

Chinese national oil companies (NOCs) are not mere puppets of their political masters. Whilst adhering to the overall guidelines provided by the government they have their own commercially driven agendas.

They operate significant domestic oil production from mature onshore fields. This is their heartland and the area of their core technical expertise. Prospects for growth within China are limited and overseas investments are deemed more attractive.

Twenty years of overseas investments have seen a marked shift from a few operated ventures in conventional fields in high risk countries such as Sudan to many (often non operated) ventures in different asset classes (including deepwater and unconventional assets) spread across the globe. Overseas ventures now account for about 30% of their oil production.

By far the largest investments were made during 2009 to 2013. They have often overpaid for these acquisitions with takeover premiums significantly higher than the industry average. But the main issue for the financial performance of these acquisitions is their timing: they were made in a high oil price world with asset prices peaked.

Towards the government, the NOCs have stressed that their acquisitions contribute to China’s energy security and to their own technical expertise (helping them to achieve their long term goal to emulate the western majors). This seems doubtful. Oil from their overseas investments is traded on the global market like any other oil. As yet, the Chinese NOCs are not seen to be able to operate in different asset classes, across the globe, in the way that the majors do.

Chinese banks have been more than willing to fund the NOCs. Chinese people, with a high savings rate, have few alternatives for their money. As for other Chinese state owned enterprises: should the NOCs run into problems these problems are shifted towards China as a whole. With Chinese debt growing at three times the rate of the economy this situation is not sustainable.

The NOCs have been a key target of China’s recent anti-corruption drive. Corruption may have been no more than a welcome pretext (the government has to be seen as being tough on unpopular corruption); it can also be seen a power struggle within the party and an attempt to reign in the poorly performing NOCs (with the aim to increase their performance).

Introduction

Some 15 years ago I worked for a small and well-hidden part of Shell in Central Africa. I have fond memories of living on the shores of Lake Yenzi in Gabon where my children grew up in a world of lagoons and tropical rainforest virtually untouched by mankind. To this day I miss the human warmth of Africa.

Towards the end of my spell in Gabon we would discuss among colleagues the arrival of a new competitor in country: Sinopec. We were puzzled. How could we reconcile the stories that the Chinese were taking over Africa (if not the world) with this hapless new venture, which had trouble getting to grips (both geology and country wise) with a completely new environment? What were we missing? I do not think we underestimated them; it was expected they would work hard and learn fast (and they had money to spend). But the general view was that they faced an uphill struggle.

These days the Chinese national oil companies (NOCs) have long shifted their focus from leftover assets in Africa to other parts of the world, including North America. The growth in their overseas oil production has been phenomenal. But it has come at a price. Earlier this year Moody’s estimated that the debt of Chinese state owned enterprises (SOEs), of which the NOCs form a major part, had risen to about 115 % of China’s GDP, higher than for any other country in the world.

There are a number of questions that I want to address in this paper. Where did the Chinese NOCs invest? Did they overpay? What were their objectives to go abroad and were they met? And perhaps most of all: have the Chinese NOCs now become global energy powerhouses or giants on shaky foundations?

 

Chinese NOCs: what kind of companies?

A heartland of mature onshore fields.  Unlike some of its neighbors (e.g., Japan or S. Korea) China has a large domestic oil production. The major fields were found in the 1950’s and 1960’s. The largest field, Daqing, has produced over 10 billion barrels and is still producing close to 700,000 barrels per day. In spite of frantic efforts, later exploration has enjoyed much more limited success. In 1993 consumption overtook domestic production and since then consumption has increased fourfold (whereas domestic production has only seen limited growth). The large dependence on oil imports (currently China imports about 62 % of its oil) is a key issue for China’s energy security.

Today, China is still the fourth largest oil producer in the world. But the bulk of its production comes from very mature fields such as Daqing, which by now experience high water cuts. It is only by intense (and costly) enhanced oil recovery methods that decline can be limited. As a result China’s onshore production is not low cost, of the order of 30 dollars per barrel on average (with a marginal cost that is much higher). Western publicity of Chinese oil companies tends to focus on their overseas acquisitions but the heartlands of these companies are mature conventional fields and their core technical expertise is maximizing recovery from these fields.

Chinese NOCs operate in a different way compared to the western majors. Their preference is to do as much as possible in house (including the use of in house service companies). If this is not possible they tend to use Chinese service companies and only as a last resort (if specialized knowledge is not available in house or in China) western service companies. Activities such as logistics and catering are done in house. Their workforces are much larger than those of western firms with similar production (e.g., CNPC employs about 550,000 people).

Government owned, not government run.  Initially oil production, processing and distribution were controlled by the Ministry of Petroleum Industry (the forerunner of CNPC) and the Ministry of Chemical Industry (the forerunner of Sinopec). In the 1980’s these ministries were converted into state owned enterprises (SOEs) and they both became integrated oil companies (be it that CNPC still has a bigger focus on the upstream and Sinopec has a bigger focus on the downstream). A third major SOE was added (CNOOC, China National Offshore Oil Company) and to date these companies (generally referred to in China as “the big three”) dominate China’s oil industry. Each of them comprises a wholly state-owned holding company and a listed subsidiary for which domestic and overseas shareholders own a minority stake (e.g., PetroChina in the case of CNPC).

To date, the heads of CNPC and Sinopec are of ministerial rank in China’s hierarchy (a higher rank than the much smaller government agencies that oversee them). To date there is no formal Ministry of Energy in China. The result has been described as “ineffective institutions and powerful firms”. The NOCs are owned by the state but not run by the state. According to an IEA report, “the top executives of the NOCs are deeply connected to the top leadership of the government and the CCP (Chinese Communist Party); they must wear two hats, as leaders of major commercial enterprises and as top Party operatives. It is in the interests of both the government and the Party that the NOCs are commercially successful, and that they secure adequate oil and gas supplies. Leaders have a great deal of freedom in how they achieve these aims, and those who fulfill them have leverage in bargaining for future promotions.” An extensive overview of the structure of the Chinese oil industry can be found in a recent OIES report.

Whilst NOCs will never omit a reference to China’s national energy security it seems that their own commercial interests are as strong a driver (if not the dominant one). There is no well coordinated master plan for China’s energy policy and overseas investments. Instead there are vague overall guidelines in an opaque environment.

The limited oversight and the opaque way in which overseas assets are acquired or work is contracted out create an environment where widespread corruption is possible.

 

The early days of going out: Sudan

The early 1990’s saw a number of developments that were of key importance to the Chinese oil industry and enabled them to go abroad. At the 1992 14th congress the CCP announced it would institute a “socialist market economy with Chinese characteristics”. Deng Xiaoping, retired from his official functions and yet at the height of his influence, believed the economic benefits of capitalism could be combined with the CCP guidance of a centralized and technically knowledgeable political system. Part of this economic reform policy involved the concept of “going out” (zou chuqu), investing surplus Chinese capital abroad to gain access to foreign markets, natural resources and advanced technology.

In 1993-1994 the Chinese government relaxed domestic oil prices, improving the financial situation of the NOCs and enabling them to invest abroad.

For the oil industry going out arrived at an opportune moment. In the early 1990’s it had become clear that domestic production could no longer keep up with consumption. The absence of exploration success and the increasing maturity of China’s producing fields implied that better opportunities for investment existed abroad. The go ahead to go abroad presented a huge opportunity to Chinese companies but also – given their complete lack of experience in operating or investing outside China – a huge challenge. But their long term aim was clear: to become competitive global businesses and to emulate the western IOCs.

Initially they started out as operators in a limited amount of countries (e.g., Sudan and Kazakhstan) with a relatively high political risk. At this time Chinese NOCs still lacked the financial muscle that they enjoyed later on and they had little choice but to go for these risky areas.

The largest of these ventures is the CNPC development of the Southern Sudan oil fields. It is also the one that has received by far the most attention in the western media. It has become the defining story for China’s investments in Africa, generating considerable reputational damage. Luke Patey’s “The new kings of crude” gives a well documented and balanced overview of CNPC’s Sudan venture (the remainder of this section is mostly based on it). It also paints a fascinating story of the pain of Chevron’s geologists (after years of hard work and exploration success having to leave the country for political reasons), the substantial achievements in development of the Chinese (establishing oil production and export in record time) and the difficult choices that the Chinese subsequently faced (with Sudanese leaders interested in power rather than their people’s wellbeing).

Throughout the late 1970’s and early 1980’s Chevron ran a major exploration campaign in Southern Sudan. It was Chevron that found the Heglig field and started the work on an export pipeline. Then things started to fall apart. An attack by Southern Sudanese rebels on Chevron’s base camp (with three fatalities) was followed by a worsening of the political environment, forcing Chevron to put things on hold. By the late 1980’s the National Islamic Front came to power and the new central government threatened Chevron to resume operations or face expulsion. A new Chevron board turned out to be less committed to the project. Making a major additional investment in a country torn by civil war was just too risky for them (also given the low oil prices after the 1986 crash). They sold their assets to a local company for a mere pittance and walked away from a 1 billion dollar investment.

During the following years domestic and small western companies found themselves unable to make significant progress (to the frustration of the Sudanese government), lacking the financial and technical clout to develop a major new oil province at a large distance from shore.

By 1995 the Sudanese search for an operator able to unlock these major finds linked up with the Chinese search for overseas opportunities. It is easy to see why CNPC was interested: significant oil had been found and although field development required a large effort it was the kind of work (development wells, pipelines) that was well within their capabilities. Chinese banks were willing to finance with loans of (up to that moment) unprecedented magnitude. With the limited choices CNPC had it was an opportunity to good to walk away from.

Oil flowing from the Southern Sudan oil fields through a 1500 km pipeline to the Red Sea by 1999 was a major achievement for CNPC. In the preceding four years they threw everything at it that they had, sending out their best teams to their most important overseas venture. They built up an entire oil infrastructure, including a local refinery. The continuing political unrest and occasional hostage taking (or worse: killing) did not deter CNPC. In any case the grueling circumstances and low safety standards were a greater danger to Chinese workers than the Southern Sudanese rebels.

During the following years Sudan’s oil production soared (to a peak of 470,000 bpd in 2007) and the CNPC Sudan venture was by far the largest producer and profit maker of the Chinese NOCs’ overseas ventures.

But after 2005 things gradually started to become more difficult. The number of incidents started to rise and the fallout of the reputational damage of the Sudan venture started to become more clear. Sudan was becoming a major hindrance in the Chinese NOCs’ overseas investments and attempts to get access to western technology. Following the large initial investments the venture gradually went into cash cow mode. Investments in enhanced recovery, needed to crank up the recovery factors, were postponed. As a result recovery factors of these fields have remained low (e.g. 23 % for Heglig, which is considerably lower than the 30 – 50 % that has been achieved for similar high net to gross sandstone reservoirs in other parts of the world). The rapid severe water cut that these fields experienced in the 2005-2010 period suggest they have been producing too fast, maximizing profit in an unstable country that was now about to split up.

For CNPC Sudan was initially a major success story. The subsequent collapse of production after Southern Sudan’s secession in 2011 has been a major disappointment, however. To this day, Sudan and Southern Sudan are arguing about pipeline fees for the transport of Southern Sudan oil through the Sudanese pipeline. The Chinese are doing their best to keep both parties happy and remain unsuccessful in doing so (in the words of a Southern Sudan oil minister: “but Jesus said one cannot serve two masters”). Political risks (both within Sudan and the reputational damage in the western world) had been severely underestimated.

 

2009-2013: overseas investment explodes

Eventually, the overseas investments of the NOCs took off in earnest in 2009. The figure below (from a presentation by SIA energy) gives an overview of Chinese NOCs acquisitions in the 2005 – 2013 period. A total of US$ 123.5 bn was spent by the three Chinese NOCs during this period, primarily between 2009 and 2013.

China paper fig1

Apart from being of a much larger magnitude the nature of Chinese NOCs’ overseas investments in the 2009 – 2013 period is markedly different from the early investments in countries like Sudan, Kazakhstan and Venezuela. There is a shift from operated assets to non operated assets, from a limited set of high risk countries to investments well spread all over the world and from primarily onshore, conventional assets to a full range of asset classes (including unconventional, deepwater and oil sands).

Several reasons lie behind this shift: the scarcity of Sudan like opportunities (large amounts of relatively low-cost, onshore conventional oil), the wish to share risk (both technical and political), the wish to not make very large investments in a single high risk country like Sudan (were the total investment eventually amounted to some $ 20 bn) and the increased importance to get access to western technology (as remaining opportunities tend to be associated with unconventional, deepwater or oil sands deposits – none of which relate to the core technical strengths of Chinese NOCs).

Landmark acquisitions during this period were the $ 15 bn Nexen takeover by CNOOC in 2013 (following a 2005 failed attempt by CNOOC to take over Unocal, in spite of putting a bid on the table that was over 10 % higher than the eventually successful Chevron bid) and the Addax takeover by Sinopec.

The question whether the Chinese NOCs did systematically overpay has generated a lot of discussion. Several papers (e.g., by Derek Scissors) have maintained that this is the case, often within the context of increasing Chinese influence in general. Many reports on Chinese acquisitions contain statements that they “again overpaid wildly” but I have seen very few systematic studies. The few I have found (e.g. a very interesting paper by Anatole Pang, one of the few papers written by someone with Chinese industry experience) were academic studies that claim they found no evidence for systematic overpaying. As these studies are based on the cost of reserves I tend to doubt their conclusions. A deal where say 2 dollar per barrel of proved reserves is paid can be a deal that is worse than one where say 20 dollar per barrel of proved reserves is paid; it all depends on development costs, tax regime, etc.

I think that looking at takeover premiums for acquisitions of publicly listed companies is the best way to deduce whether Chinese NOCs did overpay. Based on this it seems likely that Chinese NOCs did indeed overpay – by an amount of the order of 20 – 50 %. Where publicly traded companies have been acquired the premiums paid by Chinese NOCs have been hefty. Premiums paid for the Addax and Nexen takeovers were 47 and 60 % respectively; significantly above the average premium in the energy sector of about 30 – 40 %.

In takeovers of assets that were not listed they have frequently outbid competitors by significant amounts (I am not aware of any examples of the reverse).

Several factors may contribute to overpaying. Chinese NOCs may feel overpaying is necessary to overcome political resistance and to preclude a long bidding competition that may generate adverse publicity. Government approval is required and, once obtained, may be an incentive to come to a successful bid. Failed takeovers may be seen as loss of face. Government policy for the NOCs was focused on volumes and growth rather than value until recently. And finally access to funding at relatively easy terms by Chinese banks may provide less of an incentive to bargain hard for a lower price.

Nevertheless, the financial performance of Chinese NOCs’ overseas acquisitions is not so much hampered by paying more than their competitors but rather by the unfortunate timing of their acquisitions. By far the greatest amount of takeover activity took place in the 2009-2013 high oil price world. A lot of money was spent on high production cost assets, such as (Canadian) oil sands or (North Sea) mature fields that were bought at the peak of the market. These assets have performed particularly poorly in the post 2014 low oil price world.

An example is the 2012 acquisition of a 49 % stake for $ 1.5 bn in Talisman’s UK assets by Sinopec. Relatively high field decline rates, a high downtime of ageing facilities and increasing estimates of future abandonment costs limited the attractiveness of these assets already in a high oil price world (many North Sea operators have been trying to divest these kind of assets for years, with few takers). With the 2014 oil price collapse this turned into a disastrous cocktail and the poor performance of its UK assets threatened to bring down Talisman as a whole (a company already weakened by low American shale gas prices). Efforts to further divest their North Sea assets were unsuccessful and in 2014 the company was taken over by Repsol. Repsol was interested in other parts of Talisman and saw little value in the North Sea assets, especially when oil prices turned out to be lower for longer. For Sinopec a $ 1.5 bn investment turned into an abandonment-related liability within 3 years. Sinopec’s subsequent legal demand for compensation from Repsol is seen as having a very low chance of success. It is a sign of their frustration, a way to put pressure on Repsol (which values good relations with Chinese NOCs with whom it cooperates elsewhere) and stakeholder management with respect to the Chinese government.

Another example is CNOOC’s $ 15 bn Nexen takeover. Nexen, a Canadian company, is heavily exposed to high cost Canadian oil sands. Apart from its high costs, these assets suffer from being landlocked. The US blocking the Keystone XL pipeline will now result in a lower price for Canadian oil for a longer time. Even among other oil sands assets Nexen’s assets are relatively high cost and have been recently plagued by operational issues.

Many Chinese and Chinese companies lack a profound understanding of the western world (in the same way as many in the western world lack an in depth understanding of China). China should perhaps be seen as a parallel universe instead of just another country. As a result they are not optimally equipped to fully analyze the technical, political and environmental risks associated with an overseas investment.

Off course many western companies have had their share of acquisitions turned sour. But I would argue that on average they have had a better track record (paying lower takeover premiums, being more reluctant to invest in high cost mature North Sea fields or Canadian oil sands, making a better assessment of political and technical risk).

 

Recent developments (2014 – present)

From 2014 onwards overseas investments have decreased dramatically. The current low oil price environment definitely plays a role here. Profits have dramatically decreased as a result of the low oil price and write offs of previous acquisitions. Internal funding of acquisitions has become more difficult. Funding is still possible, however, and the current low oil price environment is not the only reason for the overseas investments drop.

Management of the Chinese NOCs is currently under intense pressure due to the ongoing reforms of SOEs (triggered by their poor performance) and corruption probes. A high publicity audit of $ 10 bn Angola investments by Sinopec revealed the shady deals with Sonangol through obscure companies known informally as the Queensway group. Angolan assets put on the market by western oil companies landed up (upon Sonangol exercising its preemptive rights) with companies such as China Sonangol, owned jointly by Sonangol and Chinese middlemen (but funded by Sinopec). When these assets would eventually be transferred to Sinopec (more likely so for the poorly performing assets) it would be at a substantially higher price. The Financial Times reporting on the Queensway group is one of the few cases were investigative journalism has been able to unravel the dealings of Chinese NOCs and their middlemen in some detail.

Over the last 2 years former presidents of both CNPC and Sinopec have been convicted for corruption. Many other high ranking managers have been placed under investigation or convicted. The most prominent case was that of Zhou Yangkang, who after his spell as CNPC president eventually became a member of the CCP standing committee, China’s top decision making body. Corruption may have been but a welcome pretext (the CCP has to be seen as being tough on unpopular corruption); the underlying reasons are more likely to be a combination of a power struggle within the CCP and the removal of people opposed to the reform of poorly performing Chinese SPE’s (as well as the poor performance in itself).

Knowing that unsuccessful overseas acquisitions can eventually result in convictions (be it for corruption rather than the acquisitions themselves) has made the Chinese NOCs much more cautious. Future acquisitions should involve smarter investments in quality assets, focusing on value rather than volume.

Cost cutting is now starting to result in a significant drop in domestic oil production (which still accounts for over 70 % of the total production of Chinese NOCs). 2015 is likely to have been the year that Chinese domestic oil production has peaked. By July 2016, production had dropped by more than 8 % from its peak.

 

Synthesis

On the positive side, I would consider the Chinese NOCs to be able operators for their domestic (primarily onshore, conventional) production. Average cost of the order of 30 dollars per barrel imply that these assets generate substantial profits.

Twenty years of overseas investments have resulted in equity production that is about 30 % of their total production, amounting to over 2 million barrels per day. But what else? Surely, if this were to be a true success story, it should not be just about growth.

If it is about energy security it should be noted that oil from the NOCs overseas assets is sold on the open market – and is going to the refinery that is best suited for this quality of oil and is willing to pay the highest price (and not necessarily to China). Control over the strait of Malacca (through which about 80% of China’s oil imports is transported) seems a much bigger issue here.

If it is about profitability than I feel that their record of overseas acquisitions is a mixed bag – at best. What hampers them in this regard is their record of overpaying and the timing of the bulk of their acquisitions, which coincided with the 2009-2014 high oil price world.

If it is about technical capabilities I note that, whereas they have massively invested in deepwater, oil sand or unconventional, they have done so mostly as non-operators. The technical knowledge acquired by being a non operating partner (or by acquiring a company that is subsequently run at arm’s length) is not of the same order as the technical knowledge needed to operate and grow organically. I do not see the Chinese NOCs operating e.g., deepwater fields across the globe, in a way that the western majors do. Their operated production is still primarily domestic conventional production.

I do not think that emulating the western IOCs in operating different types of assets across the world or emulating the US tight oil industry in Chinese tight oil are successful business models for Chinese NOCs. For China as a country I think it would be more beneficial to put a greater emphasis on conventional oil and gas within the Asian continent (in particular, Kazakhstan, Russia, Iran/Iraq), thus aiming at a greater share of conventional assets (closer to the NOCs technical strengths) in locations close to China that at least in part export oil by pipeline to China rather than by tanker through the strait of Malacca. Kazakhstan has so far been one of their more successful overseas investments.

The strength of Chinese NOCs (apart from their domestic production) is financial rather than technical. A western company with a similar record of acquisitions would be in severe financial trouble. Not so the Chinese NOCs: the absence of public shareholders with a short time horizon and the funding by Chinese banks imply that for them the rules of the game are different. So far, “China Inc” has bailed them out.

Their rapid growth has been fueled by profits from domestic production (in particular in a high oil price world) and by debt. Chinese banks have been more than willing to fund. Chinese people, with their high savings rate (and limited ability to move funds abroad) have few alternatives for their savings. It is for them to ultimately pick up the bill.

Chinese SOEs in financial trouble have so far been bailed out. This does not solve the problem though in the long term – it just shifts the problem upwards to the next level (which is basically “China Inc”). When evaluating the strength of Chinese NOCs one cannot look at these companies in isolation; one has to look at them as “part of China”.

In the long term, the strength of China and Easternisation are they to stay. How could it be otherwise? As Lee Kuan Yew, the former prime minister of Singapore stated: “Theirs is a culture 4000 years old with 1.3 billion people, with a huge and very talented pool to draw from. How could they not aspire to be number one in Asia, and in time the world?”

But in the short term: when will Chinese debt and the ability of China to bail out all its poorly performing SOEs hit a ceiling? At some stage pumping more debt into increasingly unattractive projects has to stop. At this stage Chinese debt is growing at three times the rate of the Chinese economy. With an increasing share of problematic loans the question is not if, but when, there will be a Chinese debt crises. Chinese that have the means to do so have now started to take their money out of the country.

The Chinese NOCs are giants on shaky foundations for a simple reason: they are part of an even bigger giant – on even shakier foundations.

China paper fig2

Posted in Geen categorie | Leave a comment

Oil Companies and Climate Change

Introduction

Climate change is real. The well documented increase in global temperature levels, the link with greenhouse gases and the again well documented rise in atmospheric CO2 levels (the main greenhouse gas) should, for all practical purposes, no longer leave any room for doubt. The vast majority of earth scientists and engineers working for oil companies do not doubt and have not done so for a long time. What has changed is the perception. From one of many problems that the world faced in the 1980’s (famine, nuclear weapons, overpopulation, “waldsterben”) this one struck us a difficult problem but one that was far away, for our children and technological progress to solve. These days we have seen that the last decade was the warmest decade on earth recorded so far. The decade before that was the second warmest. It has started to affect our lives in earnest.

The major oil companies accept that climate change is real. Furthermore they do not close their eyes for technological breakthroughs such as the dramatic decrease in price for solar panels (hopefully followed by a similar development in energy storage). If oil companies have been reluctant to invest in solar power or wind power it is not that they underestimate these technologies, but rather that they feel that solar panel fabrication is not something that can become sufficiently profitable for them or that they do not want to be dependent on subsidies. Image is important if one is to be dependent on subsidies. Oil companies are not popular; something that is unlikely to change.

The major oil companies see their strength in finding, producing and refining hydrocarbons. They expect that oil and gas demand is there to stay for several tens of years. They accept that the oil and gas industry will eventually become a sunset industry (later rather than sooner, but still). They would welcome a functioning carbon tax system. In their view it would reduce uncertainties and create a more level playing field (more so than for a system of unpredictable government subsidies and other measures to promote renewables) from which gas, the cleanest fossil fuel, could profit.

Implications for oil demand and price

Population and economic growth will add to oil demand. Increasing efficiencies, higher taxes or carbon pricing and the rise of renewables will reduce demand. How exactly this will pan out over the coming decades is highly uncertain (a key uncertainty being how quickly electric vehicles will gain market share). Statoil’s recent update of their long term scenario planning exhibits a large range for 2040 oil demand from 80 to 115 mbpd (million barrels per day). For comparison: current oil demand is just over 95 mbpd. The lower end member comes from their Renewal scenario that results in CO2 emissions in line with the target to limit global warming to two degrees Celsius. This scenario requires that the recent non-binding COP21 targets are not only met but are significantly exceeded. The higher end member comes from their Rivalry scenario where a lack of thrust and coordination result in a world where security of supply and economic growth for individual countries play a bigger role, at the expense of global climate concerns. Scenarios from other companies and organisations exhibit a similar large range for future oil demand.

Even the optimistic Renewal scenario implies that significant investments are still needed to meet a 80 mbpd oil demand in 2040. Oil field decline implies that without any activity a fields production drops on average by 8-9 % per year. Thus the oft quoted red queen analogon (“one has to keep running in order to stay in the same place” from Lewis Carroll’s Through the looking glass, and what Alice found there) remains valid. The difference between Rivalry and Renewal is that for Rivalry the oil industry needs to keep running a little faster, for Renewal it can run a little slower.

The oil industry can, and will, react to changes in demand and price in a matter of years (as we are currently seeing). It is in their best interest to do so (and they have a track record of doing so – if anything of over reacting). The energy transition on the other hand will be a matter of tens of years. I would thus argue that the energy transition is likely to result in a long term reduction in volumes but not to a reduction in price (at least not beyond the usual commodity boom and bust cycles). With relatively small changes in volumes and large swings in price it is the oil price that has by far the largest influence on oil companies’ profits.

Stranded assets and carbon bubbles

NGO’s like Carbon Tracker have made significant inroads with a theory that appeals through its simplicity. Starting with the emissions associated with a two degrees global warming limit one can derive the fossil fuel reserves (“carbon budget”) that can be burnt under such a constraint. Comparing these with the reserves of fossil fuel companies shows that a significant part of their reserves can not be burnt (“stranded assets”). With fossil fuel companies’ valuations based on these reserves this implies that their shares must be overvalued (“carbon bubble”).

Lumping all fossil fuels together, regardless of their economic value and associated emissions, is a severe simplication. I would certainly hope that the carbon budget of oil (a premium fossil fuel whose high energy density makes for instance flying possible) or gas (a relatively clean fossil fuel) can be increased at the expense of that of coal.

But the main issue I have with this theory is of a different nature: it lumps together all different types of reserves (proved, probable, possible; developed, undeveloped). In reality the value of a barrel of possible reserves, in an area where exploration may or may not prove the existence of oil, that may or may not be commercially developed, is only a tiny fraction of the value of a barrel of proved reserves that has already been developed (with significant investments for development already made). It are these low value possible reserves that run a risk of being stranded, rather than the high value proved reserves.

Proved reserves typically only account for 15 – 30 % of the total resource base of an oil company and account for 80 – 90 % of the value of a company (a detailed overview can be found in a recent IHS report). That is not surprising, given that proved reserves are either developed or in the process of being developed (for most companies project sanction is a prerequisite in order to book proved reserves). These are the assets were large investments have taken place or are currently taking place. In general, the investments related to developing a field are much larger (1 or 2 orders of magnitude) than the finding costs.

With typical proved reserves over production ratios of the order of 10 – 15 the risk that proved / developed oil reserves will turn out to be stranded is very small. Production from these assets falls far short of demand, even for a scenario which limits global warming to two degrees.

It are not the oil companies’ producing assets that are at risk but the long term continuation of their business model. But this is something that has been seen to be at risk for a long time already, be it for a different reason: the difficulties that oil companies have to replace reserves (even when spending vast amounts of money). Hence the relatively low price earning ratios of oil companies; typically of the order of 8 – 12 before the 2014 oil price drop.

A strong case can be made for carbon budgets and stranded assets in general. But those oil assets that may turn out to be stranded have been attributed a very small value. Hence I can not see a case for a bubble in the valuation of oil companies on the basis of stranded assets due to climate concerns.

All the world’s a stage and each much play a part

NGO’s like Carbon Tracker, in the words of Dieter Helm, tend to muddle up the public and the private domain. There may be a strong case for not investing in fossil fuel companies but I feel it is a case that should be based on ethical grounds rather than financial grounds (as has been done for the weapons and cigarette industries).

Aiming to reduce oil-related CO2 emissions by limiting investment in the oil industry might actually be counter-productive. Before we know it we could again enter a period of relatively high oil prices. I would much rather see high consumer oil prices due to a significant carbon tax (and using revenues for more meaningfull purposes in the OECD realm, including promoting renewables) than due to high oil prices at source level (resulting in increased revenues for Middle East producers).

NGO’s function as a lobby for renewables as well (be it a lobby that, unlike the fossil fuel lobby, has the aura of sainthood). They may well prefer direct subsidies for renewables to a carbon tax (replacing coal by gas being a very cost efficient way to reduce emissions in the short term). It is up to governments to find the right balance between a carbon tax to reduce emissions in the short term and direct subsidies to research and renewables to promote the technologies we need for a long term solution.

Oil companies have an obligation to their shareholders to maximise profits – within the limits of the law and a companies code of conduct. They are under no obligation to invest profits from oil into renewables or otherwise to contribute to solving matters that are of a public concern. It is up to governments to set the boundary conditions for the oil industry and to tax the use of fossil fuels (whether to generate revenue in general or for the promotion of renewables, to alleviate the adverse effects of fossil fuels or to discourage their use).

I feel that, from a financial point of view, oil companies will be better of by sticking to their core business and by accepting that they are likely to be in a sunset industry – in the long term. Whilst some of their assets may be stranded I cannot see the case for a carbon bubble based on stranded assets.

Black swans for the oil industry may exist but I feel they are of a very different nature. Should low cost Middle East producers change their oil policy and start to maximise volumes rather than revenue (whether due to political issues such as Saudi Arabia vs Iran tension or due to a perception that “the end of oil is near”) then this could result in oil prices being lower for longer for real. Should western governments or jurisdictions decide that oil companies should pay for the adverse climate effects of the past use of fossil fuels (something they happily allowed at the time) then this could have severe effects on oil companies profits. But let us not fool ourselves: the adverse effects of the use of fossil fuels on climate have been abundantly clear for many tens of years to governments, research institutes and companies alike. The readyness of many to demonise oil producers, whilst readily giving absolution to oil consumers, is striking.

Posted in Geen categorie | Leave a comment

Why oil prices are going up (and will continue to do so)

“a few pictures say more than a thousand words”

Oil prices have been going up over the last few months, from below 30 dollars per barrel in February to the current levels of around 50 dollars per barrel. Many short term issues of a different nature play a role here, ranging from market psychology to supply interruptions such as the Alberta wildfires. But I would argue that in the background there is a persistent upward trend, caused by the simplest of explanations: supply and demand are starting to approach a balanced situation.

On predicting future oil supply and demand no individual can produce anything close to the comprehensive analyses of organizations like the IEA or the major consultancy firms. But one can try to highlight a few processes that hopefully give some insight into the way that supply is responding to changes in the oil price.

 

Increased field decline of conventional oil fields is kicking in

Much of the publicity on the way that oil supply is responding to lower prices has focused on unconventional shale oil in the US. And indeed the rapid advance of US shale oil has been a key factor in the creation of oversupply. But for the response of supply to low prices the over 50 million barrel/day production from conventional (non-OPEC) fields plays a much bigger role. Small changes in the average decline rate of these fields result in large changes for oil supply in absolute numbers.

The figure below is taken from a recent (June 2016) Rystad Energy study and shows the average decline rate for mature non-OPEC oil fields. Up to 2015 these figures are actuals; for 2016 this is a prediction with limited uncertainty and for later years uncertainty increases.

Oil price rise paper fig1

Due to higher amounts of activity, the average decline rate in the 2010-2014 high oil price world was about 3 % only (without any activity this decline rate would have been about 8 – 9 %). Reduced activity in the subsequent low oil price world resulted in higher decline rates of about 5 – 6 %. In absolute terms: for 2016 this amounts to a supply decrease of 3.3 million barrels/day.

First of all: the fast response of existing field decline rates is due to the nature of the activities involved: workovers and infill drilling can be planned and executed in a relatively short time frame.

Secondly, the effects of reduced infill drilling in conventional fields will be felt for years to come; a conventional well not drilled in 2015 is still likely to result in lower production in 2020 and beyond.

Thirdly, decline rates are much less of an issue for OPEC fields. Production in a country like Saudi Arabia is capped by political decisions or infrastructure capacity rather than geology (hence their much higher reserves to production ratios).

So why do decline rates of conventional non-OPEC fields receive so little attention? I think this is simply because field decline has so far been masked by new fields coming on stream. For 2016, the added production from new fields amounted to about 3.0 million barrel per day, roughly compensating for the decline of existing fields. Many projects are still coming on line (especially in deepwater) that have been sanctioned in the 2010 – 2014 high oil price world.

New oil compensating for field decline is not sustainable, however. Hardly any new major developments have been sanctioned in 2015 and the first half of 2016. Over the coming years new oil from projects currently being built is gradually decreasing from 3.0 million barrels per day in 2016 to a much lower level in the early 2020’s (exactly how low will depend on when the sanctioning of new developments will resume). For 2017, a similar level of field decline (compared to 2016) is already expected to outstrip new developments by about 1.2 million barrels / day.

 

The reduction in US shale oil production is kicking in, finally

I have chosen to include a figure from a Seeking Alpha paper rather than the well known EIA figures that only give production up to the present day. Obviously the prediction beyond mid 2016 depends on future oil price (the predicted further decline is in the middle of the range for a number of forecasts from different organizations).

Oil price rise paper fig2

The time it takes for US shale oil production to respond to the low oil price oil world is much longer than often assumed. It took close to a year before peak production was reached; it took close to another year before a steady reduction (per month) of up to 100,000 barrels / day was reached. US shale oil can not take on the role of a short term swing producer (as Saudi Arabia used to do). It cannot do so timewise (it takes two years to fully respond); it cannot do so volumewise (changes in US shale oil production being too small a fraction of changes in global supply). It does play an important role, as part of a complex global oil supply system.

A shakeout is taking place in the US shale oil world with increased focus on the best plays and the best sweetspots within these plays. The Permian is emerging as the dominant play in shale oil, in the same way as the Marcellus has emerged as the dominant play in shale gas.

 

50 dollar per barrel is too low to be sustainable

The major international consultancy firms cover all parts of the energy business. Rystad Energy, a small independent Norwegian consultancy firm, primarily focuses on one part of the business only: maintaining a state of the art database of all oil fields (plus potential developments with timelines and oil prices needed for project sanction) on a global basis. This enables them to model future oil supply for different price scenarios. In this important niche they have become a world leader.

The figure below gives some these scenarios. These integrated models combine shale oil production, conventional field decline and new conventional developments (as well as more secondary processes such as exploration, end of field life and project delays). The gradual decrease of global oil supply in a constant 50 dollar per barrel scenario is primarily due to the combination of the continuation of significant mature field decline and the gradual decrease of oil from new developments over the coming years.

Oil price rise paper fig3

Unless more dramatic cost reductions materialize, these models imply that a long term price of the order of 70 – 90 dollar per barrel is needed to generate sufficient supply, anywhere near expected demand. The reason is simple: it is at this price level that many projects, of a different nature (onshore, shallow offshore, deepwater and shale oil), become sufficiently profitable to go ahead. Lower prices will eventually result in undersupply; higher prices will eventually result in oversupply (regardless of oil demand continuing to increase at its current rate or reaching a plateau).

 

Oil demand: India is taking over the role of China as a growth engine

In the meantime oil demand continues to stubbornly grow each year, at a rate of about 1.5 million barrels / day. So far any increase in efficiencies or renewables is more than offset by increasing demand in non-OECD countries. The populations of China and India (about 1.4 billion and 1.3 billion respectively) are so large that what happens in these countries simply matters most.

As the growth in Chinese oil demand abates it seems that India is taking over China’s role as the key country for oil demand growth in Asia. A recent study of the Oxford Institute for Energy Studies paints a breathtaking picture of a country where manufacturing and car ownership (and hence oil demand) are about to explode. Last year India’s oil demand grew by 0.3 billion barrels / year (compared to an average of about 0.1 – 0.15 billion barrels / year over the last decade). India is now in the position where China was about 15 years ago.

Oil price rise paper fig4

Even if India’s phase of rapid economic growth would be characterized by a greater focus on renewables this would be more of an issue for coal and gas than for oil.

One day the energy transition will take off in earnest and peak oil demand will be reached. But it will come at a slower rate than the response of oil supply to changes in oil price or changes in oil demand. The energy transition will lead to a reduction in volumes but not (necessarily) to a reduction in price. For an oil company volumes matter. But price matters much, much more.

 

 

 

 

 

 

 

 

 

 

Posted in Geen categorie | 2 Comments

The ill-fated gas strategy of the majors

Introduction

Major international oil companies have gradually shifted focus towards gas; to the extent that they are now sometimes jokingly referred to as Big Gas rather than Big Oil. For companies like Shell or BP gas now comprises more than 50 % of their total production.

Around 2010 this shift to gas still appeared to be very attractive. An expectation of continued high prices and demand growth (in particular in SE Asia) resulted in project sanction for a number of LNG projects, mostly in Australia. The 2011 IEA report “Are we entering a golden age of gas?” reflected the industry thinking at the time. The question mark in the title of the report was not taken too seriously; the rest of the report was.

To some extent this was done out of necessity rather than out of choice; replacing oil reserves had become increasingly difficult. Gas reserves are more accessible and have a wider global distribution. Cleaner gas was expected to take away market share from coal due to environmental concerns. As a transition fuel it should allow the majors to continue to grow without having to dramatically change their business model.

The strategy to move away from oil has now run into problems. Gas demand forecasts have been reduced. The onset of gas oversupply resulted in a dramatic drop in Asian gas spot prices in 2014. The subsequent, unrelated, drop in oil prices (resulting in lower gas prices for gas sold on oil-indexed contracts) exacerbated the situation for gas producers. At the same time a number of LNG projects, which have experienced large cost overruns, are about to come on the market. For the coming years the supply demand balance for gas and LNG looks worse than it does for oil.

In addition to this typical boom and bust cycle (be it one with what now looks like a prolonged bust) there are a number of more fundamental reasons why I feel that gas (and in particular high cost LNG) is systematically less profitable than oil and why the strategy of the majors to increasingly focus on gas is ill-fated:

  • Gas is a global free market; oil is not (and hence oil trades at a premium)
  • Gas is expensive to transport. Gas transport cost is often higher than the cost of feed in gas
  • Gas faces stronger competition than oil
  • Shale gas is a stronger competitor to gas than shale oil to oil

 

Gas is a global free market, oil is not.

Oil prices are higher than what they would be in a global free market. OPEC may be an organization whose members are often not able to reach an agreement but even a poorly functioning cartel is better for oil prices than no cartel.

Low cost Middle East producers are not producing to their full geological potential, whether due to political instability or due to a policy to maximize revenue rather than volume in the long term. As a result their reserves over production ratios are relatively high. Additional oil can still be developed in a country like Saudi Arabia at a cost way below that of deepwater oil or shale oil. They chose not to do so as gaining a substantial amount of market share will result in a much longer period of low oil prices than merely defending market share.

For gas, there is no such thing as a gas OPEC. Russia may not produce gas to its full potential but its role in gas markets is a far cry from the role that Saudi Arabia has played in oil markets for decades. The painful last 2 years in oil markets are the long term normal for gas markets.

Gas transport is expensive (especially when it requires liquefaction), oil transport is cheap

Due to its low energy density, gas is much more expensive to transport than other fossil fuels. Transport of gas requires pipelines (for shorter distances) or liquefaction (for longer distances). It is especially LNG that incurs high costs. Only 30 % of LNG cost is related to feed-in gas; the bulk of the cost is related to liquefaction, transport and regasification. The total cost of transporting gas in the LNG chain is at least twice the cost of transporting via pipeline.

In any country where sufficient other sources of gas are available (whether conventional or unconventional) that can be transported by pipeline LNG faces an uphill battle. In Europe, US sourced LNG will have difficulty competing with lower cost Russian gas. In China, Australian sourced LNG will have to compete with Russian sourced gas and (in future) with locally sourced shale gas. LNG is the high cost gas that faces the most pain in periods of oversupply and low gas prices (equivalent to oil sands or Arctic oil in the world of oil). It used to be profitable – at a time when it functioned as a niche gas supplier to countries (e.g., Japan) that had no alternative options.

Gas faces stronger competition than oil

The most important use for oil is transport. Alternatives for oil are less readily available in the short term. Even for light vehicles a transition to electric vehicles will take considerable time. It remains to be seen if (and when) the use of electric vehicles can compensate for increased road transport on a global basis. Decades of increased fuel efficiency, for any form of transport, have as yet not resulted in peak oil demand. For heavy vehicles, airplanes and shipping a transition is even more difficult.

The most important use for gas is power generation, where coal and renewables are strong competitors. The low cost of coal remains a strong advantage, limiting the rate at which especially non OECD countries will move away from coal. Reduced costs and climate concerns result in renewables now making significant inroads – which is more of a concern for gas than for oil.

The only place where gas has a high and increasing share in power generation is the US. This is solely due to low cost shale gas – which does not help the majors in any way. In Europe gas is being squeezed in between coal (which still enjoys significant political support in Eastern Europe) and renewables. European gas demand is 20 % lower than what it was a decade ago.

Countries like China and India have so far chosen cheap coal for the bulk of its power generation. By now, should  they want to start reducing the share of coal, they may move straight to renewables, bypassing gas. The IEA now expects gas to be responsible for only 8 % of Chinese power generation in 2040, up from the current 4 % but still way below a global average of about 23 %. Gas (and especially high cost imported LNG) is simply not the best compromise between cost, emissions and energy security for a country like China.

The majors are not making much progress in selling gas as a transition fuel. It is cleaner than coal and yet it remains a fossil fuel and methane emissions are subject to increasing public scrutiny.

Shale gas is a stronger competitor to gas than shale oil to oil

The BP long term scenarios have shale gas providing for about 25 % of the total gas supply on a global basis in 2035. Shale oil is only expected to provide for about 10 % of the total oil supply by that time. Other scenarios, such as those provided by the IEA, paint a similar picture.

The situation in the US, the only place where shale oil and gas are mature industries, provides the background for this. US shale gas is firmly established at the lower end of a gas cost curve. Since 2009, when shale gas took off in earnest, it has completely outcompeted conventional gas in the US. Shale oil, on the other hand, faces more of a struggle. Shale oil projects have a significant cost range but on average US shale oil is situated in the middle of the global oil cost curve.

Given the knowledge and efficiency of the US oil (service) industry any non US shale oil will be at a higher cost (and will struggle to reach the level of activity needed to bring down costs and establish sweet spots). Long term scenarios like those created by BP or the IEA expect non US shale gas to have a higher chance to take off than non US shale oil. For any place in the world where shale gas can overcome the technical issues in the early phase as well as public acceptance issues it may outcompete conventional gas (and in particular high cost LNG) as it has done in the US. This may be unlikely to happen in Europe but has a real chance of happening in China or Argentina.

The current LNG oversupply is more severe than the current oil oversupply

 For oil, the difference between supply and demand over the last two years has not exceeded 2 million barrels / day (close to 2% of the total production). By now (June 2016) supply and demand are starting to approach a balanced situation. In anticipation of a further reduction of supply (related to the investment cuts over the last two years), oil prices have started to pick up and the lowest prices seems to be behind us.

For LNG, the length and intensity of the bust period of low prices is expected to be much more severe. By 2014, a well supplied LNG market became a buyer’s market, resulting in a significant drop in e.g. Asian spot prices in 2014. What is the most worrisome at the moment is the number of LNG projects that are now coming on the market. Global LNG exports are increasing from 233 m tonnes (2014) to 306 m tonnes (2016). The bulk of this increase comes from Australian projects – all destined for Asian markets that are at this moment hardly growing (by about 2 % per year only – much less than foreseen 5 – 10 years ago when these projects were sanctioned). This oversupply is of such magnitude that it is likely to lead to a prolonged period, at least to 2020, of LNG oversupply and low prices. Following large costs overruns, a recent Australian LNG project such as Gorgon runs the risk of becoming one of the worst projects from a financial point of view in the oil and gas industry since a long time. US LNG projects are having a significant cost advantage compared to greenfield Australian projects due to lower construction costs and lower costs of feed in gas (more than compensating for longer transport); with e.g. Japan delivery costs estimated to be about $ 11 / MMBtu versus $ 14.5 / MMBtu.

For the future, much will depend on how much output has been tied to the oil price (and of course how oil prices will evolve). In the present world of low oil prices gas spot prices tend to be relatively close to those of long term contracts. I would expect the outlook for oil prices to be better than that of LNG spot prices. More than 75 percent of all Asian gas import are priced at levels contractually linked to oil prices (versus less than 50 percent of European gas). In the long term there is a tendency to move away from oil linked prices to spot prices or hybrid pricing. At this stage Asian consumers are reluctant to sign any new oil-indexed NLG contracts. Part of the contracts that are being signed go to portfolio players rather than destination specific end users.

Europe is unlikely to absorb excess LNG on a significant scale. Gazprom is unlikely to cede market share. As a low cost producer they can undercut on price and they have significant spare capacity (cost levels of about $ 3.5 / MMBtu for existing Russian spare capacity, $ 5.5 / MMBtu for incremental Russian capacity versus approx $ 8 – 10 / MMBtu for US LNG). European LNG needs strong political support (for environmental or energy security concerns) in order to be successful.

 

Concluding remarks

In the long run, gas seems to be systematically less profitable than oil. In the short term, the current low LNG prices are expected to last a lot longer than the current low oil prices. In hindsight the majors would have been better off accepting a shrinking business with a more limited focus on gas and a much more limited focus on high cost LNG.

For a different perspective (but arriving at quite similar conclusions) I would recommend Karel Beckman’s paper on the 2015 World Gas conference. It contains some interesting observations on the oil and gas industry’s groupthink.

 

Gas paper 21

 

Gas paper 22

Gas paper 23

 

Posted in Geen categorie | Leave a comment