The Era of Cheap Oil is Over

Download PDF

I’ve heard some very smart people speculate that oil prices will decline as new supply comes online – mainly related to shale oil. This just isn’t the case.

– Harris Kupperman, Adventures in Capitalism


Harris is 100% right. And I’m going to prove it to you.

Side note: This is Part 1 of my 3 part Unconventional Oil theme (you can read more about my themes here and my investment approach here). Over the coming weeks I’ll publish Part 2 (a playbook of how to evaluate unconventional oil opportunities; update: published here) and Part 3 (a specific idea within this theme that I suspect will make investors a lot of money over the coming years; update: published here). To get these notes the second they are published, enter your email here:

But now onto Part 1 – understanding the forces driving a sustainable bull market in oil.

Let’s start high-level. There are 3 factors that influence the price of oil:

  1. Demand – driven by per capita consumption and population growth
  2. Supply – driven by new discoveries and spare capacity
  3. Inventories

Diving into each and doing simple math will lead you to the same indisputable conclusion I have come to: the era of cheap oil is over.


Before I lay out my thoughts, I think it’s best to start with the thoughts of someone much smarter than me – John Hess of Hess Corporation. In a 2009 speech, he offered 3 main points (source):

  1. Fact No. 1: Eighty-five percent of the world’s energy comes from hydrocarbons. While renewable energy will be needed to meet future energy demand and contribute to reducing our carbon footprint, hydrocarbons will fuel the world’s economy for decades to come. Renewable energy does not have the scale, timeframe or economics to materially change this outcome.
  2. Fact No. 2: Once the economy recovers, oil demand is projected to increase by 1 million barrels per day each year, as world population grows from 6.8 billion today to 9 billion by 2050. The introduction of higher mileage standards in the U.S. and the gradual phasing in of electrical power into automotive drive trains will only moderate growth in automotive fuel demand. That is because nearly one billion vehicles on the road today could grow to approximately two billion vehicles in the next 30 years. Keep in mind: The U.S. has 1000 cars for every 1000 people; China has 10 cars per 1000.
  3. Fact No. 3: Supply. We are not running out of oil. The issue is not our endowment of oil resources, it is the world’s production capacity. Additions from exploration last replaced annual production in 1987. The easiest oil has been discovered. Costs are increasing for new barrels, where wells can be drilled in water depths of over one mile to targets up to six miles deep, and discoveries can take over a decade to develop. Oil field declines are running at more than 5 percent per year. That means we have to add at least 4 million barrels per day each year just to keep production flat. Yet non-OPEC production is in the process of, if not peaking, reaching a plateau. The U.K. Energy Research Centre just published a report that there is a significant risk that worldwide production of conventional oil could peak before 2020 and enter terminal decline. If we do not act now, we will have a devastating oil crisis in the next 5-to-10 years.

Factor 1: Oil Demand

On the demand side, the one thing in John’s speech that immediately jumped out at me was this: oil demand is projected to increase by 1 million barrels per day each year, as world population grows from 6.8 billion today to 9 billion by 2050.

There are two points here. The first is that – absent a recession – oil demand will grow at a rate of 1MM barrels per day each year. Looking back at the past 3 years of data (a relatively stable period ex-Europe) demonstrates that John is right – global oil demand has increased an average of 1.2MM boe / d per year. Why has that happened? Because of the billions and billions of people in the developing world striving for a higher standard of living. As more people join the middle class, cars replace bikes and oil consumption increases. It’s that simple.

To provide some context around this burgeoning demand growth, consider this:

US daily oil consumption = 19MM barrels
US population = 318MM
Per capita consumption = 0.06 barrels / day

China daily oil consumption = 10.2MM barrels
China population = 1.365BN
Per capita consumption = 0.0075 barrels / day

A person in the US consumes 8x the amount of oil per day as a person in China. Compared to an Indian, it’s 20x. That is an absolutely staggering difference.

Truthfully, the US isn’t a great comparison to these developing economies – the wealth gap is just too wide. But I think Brazil is a useful comp. So let’s assume that the per capita consumption in only 3 countries – India, China and Indonesia – rises to match Brazil. What would that mean to oil demand?

It would mean additional demand of nearly 30MM barrels of oil per day, 30% more than the world currently consumes.

Increase Assuming Brazil-Level DemandSource, Buyside Notes:

This point needs emphasis: if only 3 countries experience Brazil-level demand, the world will need to produce 30% more oil to meet that demand.

The second point I want to highlight from John Hess’ speech is on worldwide population growth. And to this point, he is being conservative. If you follow the work of Hans Rosling, you will understand that current demographic trends indicate the world population is a lock to hit 10BN people (to see why, watch minutes 10:30 – 13:00 of this video). This isn’t fantasy; it’s a virtual certainty.

So what happens to oil demand when you add 3BN people to the world’s population? Well, if you assume exactly the same per capita oil consumption as exists today (meaning no improvement from a growing middle class, no per capita GDP growth, no Brazil-level demand), the world will see a demand increase of 36.1MM barrels of oil per day. It’s a near certainty that the world needs 40% more oil. 

But what if you combine these two trends – a population increase to 10BN and a Brazil-level per capita increase in India, China and Indonesia?

If you experience both a Brazil-level demand increase in India, China and Indonesia along with 3BN extra people, daily oil demand increases by 67MM barrels of oil per day. That is over 70% more oil than we consume today. And the implications for oil prices should be obvious.

Now you can begin to appreciate the last line in John Hess’ speech: If we do not act now, we will have a devastating oil crisis.

Factor 2: Oil Supply – Overview

Over the long term, changes in oil supply are driven by a four main factors (I’m intentionally ignoring short-term supply disruptions):

  1. New discoveries
  2. Spare capacity
  3. Decline rates
  4. Improved extraction technologies

Before getting into the factors that change supply, it’s important to understand the make-up of our current supply:

Crude oil production is heavily concentrated in a small number of countries and a small number of giant fields, with approximately 100 fields producing one half of global supply, 25 producing one quarter and a single field (Ghawar in Saudi Arabia) producing approximately 7%. Most of these giant fields are relatively old, many are well past their peak of production, most of the rest seem likely to enter decline within the next decade or so and few new giant fields are expected to be found (source).

What stands out here is just how important giant oil fields are to the world’s supply (a “giant” is defined as a field able to produce at least 500 million barrels of oil over its lifetime). Giants are responsible for over 60% of global oil production:

Giant Oil Field ProductionSource, p 86:

Factor 2(A-1): Oil Supply – New Discoveries – Giants

As demand continues to rise, new giants have to be discovered to meet the demand. Unfortunately, giant oil field discoveries peaked in the 60s and are becoming more and more scarce:

Giant Oil Field DiscoveriesSource, p 83:

Not only are the number of discoveries down, but the size of each new discovery is smaller:

Giant Oil Field Discoveries - TableSource:

As you can see, there hasn’t been one discovery since 1980 that has – or is expected to have – a production capability of more than 1MM barrels of oil per day.

What’s clear is that the giants are getting smaller.

A final note on this subject: Discovery does not equal production. As Harris points out in his blog post on Kashagan – by far the largest of the recent giant discoveries – execution issues have pushed out the production timetable another few years. This is a field that was originally projected to come online in 2009 – now it’s not expected to produce until 2017!

This is a fiasco of epic proportions. And it has significant implications for other major oil projects – if you don’t think engineers everywhere are scrambling to re-evaluate the cost of their own project you are badly mistaken. And herein lies perhaps the biggest gap in understanding between financial analysts and oil producers: a true appreciation for the complexity, cost and calendar of major oilfield developments. And because of this under-appreciation, oil continues to catch a bid as supply surprises the market to the downside.

Factor 2(A-2): Oil Supply – New Discoveries – Shale

Now it’s time to address the biggest energy myth ever propagated by CNBC: “energy independence” driven by the shale boom. It’s a dangerous lie. (Side note: lies can generate fantastic investment opportunities because they create gaps between facts and the market consensus. Investing in these gaps is what Wall Street legend Michael Steinhardt calls “variant perception investing”.)

Conventional wisdom says that shale oil will allow US production to exceed that of Saudi Arabia and US imports will shrink to zero. The reality is vastly different. To understand why, you first have to understand the lifecycle of a shale play:

Shale LifecycleSource, p 13:

The last two bullets are very important: producers want to recapture their investment early and go after the most prolific areas within a play first. Those wells ramp quickly but experience dramatic decline rates – 90% or more in the first 5 years. By then, the play is already middle aged.

To make this concrete, think about shale oil this way: once a producer puts the drill bit into the ground, they are running on a treadmill. A very, very fast treadmill. To keep up, they have to invest huge amounts of capital to replace lost production from sky-high decline rates. First year decline rates in North American shale plays range from 50% – 67%:

Shale Decline Rates
Source, p 6:

Visually, a typical shale oil type curve looks like this:

Shale Decline CurveSource, p 31:

You can see how quickly the production in a typical shale well declines. The fall-off is so dramatic that the chart takes the shape of an “L”.

With that as a backdrop, I’m going to go out on a limb and make the case that the great shale boom is already over.

For simplicity, let’s start by modeling 1 drilling rig that completes 1 well every 3 months – I am doing this to show how production scales if rig count / investment stays flat. Here’s what the average daily production curve looks like in 1 year:

Shale Decline Curve - 1 YearSource, Buyside Notes:

And here’s what that curve looks like in 10 years (same assumptions: 1 drilling rig, 1 new well brought online every 3 months = 4 wells per year):

Shale Decline Curve - 10 YearsSource, Buyside Notes:

It might be hard to notice the trend in the above graph, so here is the data in table form:

Shale Decline Curve - 10 Years, TableSource, Buyside Notes:

What should immediately jump out is the Gamma column – that is the velocity of change. What the table shows is that – assuming a stable rig count and consistent investment – the shale market will experience a very large initial surge of production in the first two years followed by much smaller growth in later years. So if producers want to sustain the very high levels of growth of the first two years, they have to drill more wells and invest increasingly more capital than they did at the outset. Four wells drilled in Year 1 boosts production by 335 barrel per day; four wells drilled in Year 10 only increases production 110 barrels per day.

If shale production increases are to continue as they have in the recent past, the number of wells being drilled is going to have to increase every year. And more wells means more drilling rigs. So the key to shale production growth is rig count growth. And here’s what the rig count data is telling us:

US Rig Count


What you’ll notice when you look at the graph is the massive shift of rigs from natural gas to oil starting in 2008. This was the beginning of the shale oil boom and resulted in a million barrel per day increase in North American production in both 2012 and 2013. What you’ll also notice when you look at the graph is that from 2Q12 to 2014, the oil rig count has stayed fairly constant. Using the decline curve logic I presented above (more rigs required to achieve the same production growth given decline rates), the conclusion should be obvious:

The great shale boom is already over.

Now, let me be clear: North American oil production is going to continue growing in the coming years. But it’s becoming apparent that growth will come at a much slower rate. And I believe this will surprise most investors; those that don’t understand the relationship between decline curves, investment and rig count will overestimate supply and underestimate the price of oil.

There are a few other points worth making on the topic of “energy independence” (source):

I’ll close this section with one final point from Amrita Sen of the Oxford Institue (source):

Factor 2(B): Oil Supply – Spare Capacity

Spare capacity is the volume of production that can be brought on within 30 days and sustained for at least 90 days. Basically, it’s what’s available but not “turned on”. Because non-OPEC countries are price takers / net importers, worldwide spare capacity = OPEC spare capacity. Generally, 2.5MM boe / d of OPEC spare capacity is considered the lower limit of what the oil market is comfortable with. Current spare capacity is 4.65MM boe / d (source) and recent longitudinal data can be seen here (note: the bright blue columns are estimates):

OPEC Spare Capacity


Despite expectations for nearly 7MM boe / d of spare capacity in 2014, the latest data from April 2014 shows total OPEC spare capacity of 4.65MM boe / d; if you take out Iraq, Nigeria, Libya and Iran, spare capacity is just 3.41MM boe / d. The fact that spare capacity isn’t meeting market expectations provides further indication of a tight oil market globally.

Factor 2(C): Oil Supply – Decline Rates

Oil has one thing in common with every other finite resource / non-renewable commodity: depletion. Currently, the global supply of oil depletes at a bit over 5% per year. Think about that… just to stay flat with last year’s production, the world needs to bring on 4.5MM additional barrels per day each year. To put that in perspective, the entire supply growth from US shale last year was 1MM barrels per day. When you frame depletion this way, you start to get a sense for how difficult it is to offset the impact of natural declines each year.

Factor 2(D): Oil Supply – Improved Extraction Technologies

Unknowable unknowns exist in every market. In the oil market, perhaps the biggest unknown is the impact of new extraction technologies and enhanced oil recovery (EOR) methods on future supply. What I’ve seen from researching oil companies over the last couple of years is that EOR methods can attenuate decline rates but aren’t really capable of providing a big new wave of production growth. EOR programs provide producers – especially in high decline shale resource plays – with low investment / high return on capital opportunities, but they can’t solve our supply problems.

One area that I’m watching closely is heavy oil. Heavy oil comprises 50 percent of known oil resources but represents just 10% of worldwide production. Any technology that can crack the code commercially would be a complete game-changer for the oil market and unlock 8 trillion barrels of additional production. So far, there have been lots of attempts (Petrobank’s THAI technology is a notable example), but nothing to date has been able to achieve commercial success.

Factor 3: Inventories

Current OECD inventories are at their lowest levels since 2008. Why? Because of the “surprising robustness” of demand in the U.S. and other developed nations (source).

OECD InventoriesSource, p 18:

But the story today is not that OECD inventories are at critically low levels, or that inventories are falling rapidly. Rather, the point is that inventories are not building. On an absolute basis, OECD inventories are at similar levels as the 2006 – 2008 period (2.5 billion barrels as of March 2014). But one must remember: this has been a period of below-trend growth. What will happen when worldwide growth picks up?


There are two additional, inter-related characteristics of the oil market that make me extremely bullish over any timeframe – short or long-term. They are:

  1. The oil production cost curve.
  2. The “put” held by producers.

When you talk to industry insiders, a consistent message emerges: high oil prices are required to bring on new supply. Let’s do a mental exercise and assume oil drops to $70 in the short term. What happens in that case? Within a year, oil would be at $120. Why? Because at $70 per barrel, shale oil doesn’t make any money. And as soon as oil drops, producers will lay down their rigs because they aren’t economic. Given the extremely high decline rates in shale, when drilling stops the impact on production is immediate. Any significant short-term drop in oil creates a large drop off in production, pressuring oil prices upwards. This thesis was confirmed on a Core Laboratories conference call in April 2013:

David Demshur – Chairman, President & CEO: […] our view is that when we looked at a lot of the unconventional plays, you’ve got now WTI that has slipped below $90. For us, that’s kind of a benchmark where we feel outside of the sweet spot in the Bakken, in the Eagle Ford, and then the Niobrara that it’s very difficult to reach a reasonable return on investment for our clients.

James West – Barclays: Okay. That’s pretty interesting given your view that $90 is kind of the threshold level would suggest that WTI is not going to stay there for very long or below $90 very long?

David Demshur – Chairman, President & CEO: Yeah, well. I don’t know how long it’s going to stay there, but the $90 WTI, the level that is necessary for reasonable return is not our view, it’s our client’s view.

This is what I mean by the “put” that producers hold. If oil falls below a producer’s cost curve, they will scale back their operations, supply will decrease and prices will jump back up. And it’s not just the shale producers that hold this “put” – it’s every single oil company. To understand the level each producer will exercise their “put”, you just have to look at the cost curve:

Oil Cost CurveSource, p 32:

I would note that the above chart doesn’t factor in social contracts; for example, many believe Saudi Arabia requires $100 oil to sustain their promised public spending (source).

So what does all of this mean for investors?

When you pair certain demand growth, driven by per capita consumption and predestined population growth, with lower-than-necessary supply growth, driven by fewer and smaller giant discoveries, mega project delays and the decline / depletion treadmill, the conclusion seems certain:

The world is going to be short oil. 

Now marry that conclusion with the oil producer “put” and the answer reveals itself:

Expensive oil is here to stay. Any dip is a buying opportunity. And every financial model should assume $90 as the new floor.

Post Script: There are a few future developments that would make me reconsider my position: (1) new technology that can extract heavy oil economically (this would set a new “floor” in oil), (2) electric car sales approaching 10% of total car sales (assuming worldwide vehicles stay at ~1 billion), and / or (3) solar becoming a meaningful source of energy (note that Jeremy Rivkin thinks we’re ~25 years from a solar age; see min 15 here:

One Response
  • TEs Reply

    Well…any thoughts here? WTI around $80 and there’s a lot of panic among small and mid-caps…

Add Comment Register

Leave a Reply