This paper attempts to rationalize recent oil price strength and perspectives for the remainder of the year. Post-trough recovery and demand-led strength have been substituted by geopolitics as prime oil price driver since early Q2 18
After the broad based commodities recovery of 2016/17, oil price strength has persisted in an almost unabated fashion. At currently USD 74/bbl, Brent Crude is now up more than 150% from the 2016 trough of USD 29/bbl.
In order to gage a better view on these latest movements, we try to disentangle the influence of diverse factors on oil prices. We first estimate the log of Brent oil prices in function of the log of a world GDP now index (based on Goldman Sachs world GDP now-casting index), the log of world oil supply (data from the US Department of Energy), the log of net long positions on the Brent futures and the log of the Dollar index (DXY), plus one constant, all on a monthly frequency (from January 2001 to January 2018, data on world oil supply beyond that date are still missing).
We use an error-correction model1 allowing the identification of a long-term and short-term equation. We can interpret coefficients in the long-term equation as elasticities.
It means that 1% increase in the world GDP triggers a 1.7% increase of oil prices, while an increase of the same extent of world oil supply triggers a decline by -2.6% of this price. An increase of 1% of net long positions in Brent futures triggers an increase by 0.7% of oil prices, while an appreciation by 1% of the Dollar index induces a decline by 2.4% of oil prices.
We compare our theoretical model with observed data and get a satisfying result (R² = 0.85) and extend our estimate to May 2018 by assuming that the m/m variation of the world oil supply is the same in 2018 as in 2017 for the missing data (February to May 2018).
The residual of the long-term equation could be associated with the geopolitical factor, which we tried to put in this equation via the inclusion of a weighted average of a geopolitical risk index (Saudi Arabia, Russia, Mexico, Malaysia, Indonesia…) but it was not significant.
Looking at the contribution of each of our factors to the fluctuations of modeled oil prices we can see that fluctuations of the Dollar and speculative (or momentum) factor contribute the most to the variations observed in oil prices. The contribution of real demand embodied by world GDP growth and the contribution of world supply growth seems to be marginal . However, the momentum or speculative factors and the currency factors seem to be amplificatory of shocks in demand and supply in the real side of the economy.
At this stage, the market has moved from demand driven to possibly supply constrained.
Post first trough (USD 50/bbl 01/16), the recovery from H2 2017 was driven by stronger than expected demand on the back of synchronized global economic growth.
Earlier this year, we upgraded our global GDP forecast by 10bps to 3.3% for 2018.
Despite our expectation of a Q2 18 soft patch, our full year view in terms of growth in unchanged, for sustained strength of demand.
Figure 4 Brent Crude (USD/bbl)
We have envisaged the possibility that the residual of our long-term equation represents geopolitical risk as this is the only important factor influencing oil prices, which is not included in it. It seems that this factor has played a bigger role over the last year.
To verify this assumption, we estimate another equation between January 2007 and May 2018 with the world GDP, the geopolitical risk indicator and the Dollar index.
The performance of the model is satisfying and sign of coefficients as theoretically expected (see Figure 5). Over this period, the geopolitical risk index is significant and more visible especially in a more recent time.
We can see that a lower contribution of fluctuations of the Dollar (i.e. the recent re-appreciation of the USD) is probably one the main driver of increase in oil prices, beside the rebound of world growth and the
rise of geopolitical risk (see figure 6).
Oil price strength in Q2 to date has indeed been driven by geopolitics, in particular heightened tension in Syria and in the Middle East, and subsequently the US withdrawal from the Iran nuclear deal.
While there was no direct supply impact on supply from Syria, the possible loss of production from Iran could range from 0.2m to 1mbpd.
Figure 7 Brent oil prices (contributions, y/y, %)
Oil price strength in Q2 to date has indeed been driven by geopolitics, in particular heightened tension in Syria and in the Middle East, and subsequently the US withdrawal from the Iran nuclear deal. While there was no direct supply impact on supply from Syria, the possible loss of production from Iran could range from 0.2m to 1mbpd. There is unlikely to be a full loss of production from Iran: Over the short term, exemptions will mitigate to a degree. There is stiff opposition to the US withdrawal from the agreement.
An at least partial upholding of the deal is not outside the scope of possibilities. Compliance levels with sanctions might be lower this time than previously where there was multi-lateral buy-in. However, the shortfall from Iran comes at the same time as prospects for further shortfalls from the Venezuelan economic and political crisis, likely in the region of 0.5mbpd. Altogether, in a realistic scenario, the market could lose about 1mbpd of oil supplies going forward.
There is potential mitigation to the loss of supply. OPEC could step up production, notably Saudi Arabia. After the supply cut agreement, there are about 2mbpd of spare capacity. Saudi ministers have confirmed that there is available capacity. But Saudi Arabia is unlikely to act on its own; any increase in production would have to be agreed within OPEC and with Russia. The question is whether Saudi Arabia is confident enough with regards to supply cuts having achieved a balanced market in a persistent manner.
Saudi Arabia denies having a target oil price. Yet, it – and other oil producing nations – clearly have a choice of long term optionality in relation to their resource, according to optimization occurs. And rhetoric, such as debate over new measures of inventory averages, suggest that supply cuts may remain in place over the short term.
We will be watching the run up to the June meeting where a revision to the supply cut agreement could occur, closely. We further believe that the Saudi Aramco IPO, albeit uncertain in terms of timing, should not be entirely ignored when it comes to considering oil prices and a possible step up of Saudi production volumes. A higher oil price will have a greater impact on pricing than temporarily higher volumes.
The US government has entered into talks with international oil majors in order to gauge scope for them to increase production. Sanctions coming into place gradually may allow other producers to increase output. This concerns conventional resources and shale oil.
Furthermore, higher oil prices incite natural mitigation through increased production. Certain companies in the shale sector claim profitability for their best projects at USD 25/bbl. While this is certainly the very low end, tier one share projects are now profitable around the USD 45/bbl level. Second and third tier projects can be justified around the USD 65/bbl level.
This leaves ample margin for production to be stepped up in response to price signals. US shale production would likely the first to increase. Note, though, that the shale industry is encountering capacity constraints which will limit production increases. The most important issue is pipeline capacity in Mid Texas. As production from the Permian Basin, the most important production region in the US with 55% of all active rigs in the country, has increased by 60% over the past two years, pipeline capacity expansion has not kept up. There are also reports of tightness in the supply chain for production, notably sand, needed for fracking. However, there is probably room for about 0.3-0.5mpbd of additional production for 2018 in our view. Additional pipeline capacities of 1.9mbpd are due to get commissioned in H2 19. That will lift US shale production solidly above 11mbpd.
Other fundamental data has been soft. While the inventory overhang has mostly cleared and demand remains strong at likely just short of 100mbpd in 2018– inventories currently stand 10mb below the five year average – recent data has shown surprise increases at some occasions.
At the same time, US production continues to increase. The EIA has recently increased its production forecast to 10.72mpd for 2018. Rig count is up 11% ytd. The great earnings recovery – earnings by the oil majors have grown 42% y/y on average in Q1 18 – increases the potential for investment into new projects by virtue of greater availability of cash. The industry is more disciplined than in the previous cycle as far as capex is concerned, but we do think that pressure for volume growth will mount.
There is also risk of a negative demand effect as a result of much increased oil prices. Demand elasticity tends to be highest in emerging markets. We note the IEA’s 40kbpd reduction in its 2018 oil demand growth forecast. We note that financial markets’ views on the oil price are mixed. In particularly, managed money net long positions have decreased since early 2018. The historic correlation between oil futures and net long positions and has been broken since early Q2 18. We calculate an average oil Brent price of USD 68/bbl ytd. Based on GDP growth of 3.3% for 2018, about 0.5mpbd supply reduction and a 2.5% USD appreciation, our econometric model suggests a Brent price of USD 72 on average for the whole of 2018.
However, we are very aware of the bulk of this being driven by the strong increase in net long positions on a 12 months average basis and us being only part into the year 2018, while such positions have in fact decreased since early 2018 as mentioned above.
Figure 9 US crude production and inventories (thousands of bbl/day)
For our central case, we assume a return to the 24 months average as far as net long positions are concerned. Interestingly, when taking into account only the increase in net long positions over Q2 18, which we believe to be the quarter where oil prices are predominantly being driven by geopolitics, our model fully explains the current price of Brent Crude.
All the while we are not in any way akin to suggest that speculators or managed money move the oil price, we believe the conclusions from our model confirm that the market is being driven by concerns over geopolitically induced supply shortages (that may or not materialize).
Assuming a base case of stabilization around current levels for the remainder of this quarter, followed by normal seasonal pattern would also imply an oil price of USD 72/bbl on average for 2018. We rationalize this by a sustained high price level on the basis of uncertainty over the impact of the Iran issue and the actual impact hitting the market later in H2.
Should markets correct to the mid USD60/bbl levels in Q3 on the basis of fundamental data and/or pull-back after the very rapid recent increase –possibly driven by reducing net long positions after strong performance - this would bring the average for 2018 to USD 67/bbl.
We have included a bull and a bear scenario in order to account for variability in our underlying assumptions: 3.3% GDP growth along with 2% USD depreciation y/y and greater supply loss than in our base case – 2mpb from Iran and Venezuela combined – would yield a Crude price of USD 80/bbl.
Conversely, our bear scenario assumes 30bps of loss of GDP growth, possibly from a negative oil price impact, along with 5% USD appreciation y/y.
It further assumes that the industry would be able to fully make up for the loss of supply from Iran and Venezuela and that the US shale industry could bring an incremental 0.5mpbd to market, i.e. total supply increasing by 0.5mbpd y/y. This results in USD 67/bbl Brent.
For 2019 we assume 3.1% World GDP growth and 2.5% dollar appreciation. However, significant new supplies will come into the market as the US Permian Basin gets de-bottlenecked. Furthermore, oil prices topping USD 70/bbl will incentivize other new production. We have baked 1mpd of incremental production into our forecast, which we think is conservative. On this basis, our central estimate is for an oil price of USD 69/bbl. Companies in certain sectors are now finding themselves with materially increased input costs. Given the speed of the rise in the oil price, most likely, there will be little hedge in place. The sectors most concerned are sectors that transform crude oil, such as specialty chemicals, but also energy intensive industries such as heavy industry, manufacturing, and certain segments within machinery. Airlines, road transport and shipping will see their fuel cost increase. There will also be an impact on the automotive sector. Increasing oil prices can lead to an acceleration of substitution of vehicles, e.g. towards EVs.
For that to happen, the increase needs to be seen as persistent. Mining, a very energy intensive sector, particularly for open pit resources, faces higher fuel costs for diesel used in trucks and digging equipment. Iron ore and bauxite which feed into steel and aluminum, are particularly exposed. Most of the B2B sectors at this stage are able to pass the increased input costs from higher oil prices on to customers. This is the case for specialty chemicals and certain machinery manufacturers at the high end.
As long as pricing power persists, margins will be protected. It is the end consumer who pays for the increase in oil prices through inflation in a broader basket of goods and services. Other sectors and businesses in segments with intense competition will see their margins contract.
Several mining companies have stated they are looking for greater operational efficiencies as the sector. However, during the week commodities cycle, a lot of potential for efficiencies has been exploited and
further cost reductions will be more challenging. Pricing power in the mining sector very much depends on the tightness of the market for the commodity in question. There are segments where high cost producers will need to close down operations. Steel and iron ore are sectors of particular concern in that context.
Importantly, higher oil prices encourage substitution. Alternative processes, clean energy technologies and renewables will benefit as they become more competitive. A number of alternative processes become competitive with hydrocarbons based technologies around the USD 70/bbl level. The mature power generation technologies are already now at levels where they are competitive with fossil fuels on a stand-alone basis. Still, adoption rates tend to increase whenever oil prices rise.
Last not least, the oil and utilities sectors are seeing a significant earnings recovery and return to growth on the back of high oil prices.