Bill Gilmer, director of the Institute for Regional Forecasting
Forecasting the future is next to impossible. It is hard enough just to figure out where you are and how you got there. But I think we can find some perspective on current oil markets by applying the elementary logic of Econ 101 to the current price collapse. Throw in a few facts about the size of oil markets and how they work, apply a little informed speculation and we actually know a lot about where oil prices are and why – and even where prices are going.
First, as scary as oil markets look today, there is no analogy to 1987. The 1970s saw a series of supply shocks, brought on by turmoil in the Middle East and the rise of OPEC. As oil markets worked their way back to long-run equilibrium by bringing on vast new oil supplies, and while OPEC grotesquely overplayed its hand as a cartel, it unleashed a flood of at least 8.0 million barrels of excess capacity, equal to 13 percent of global production.
It took a decade for the world economy to absorb this oil surplus.
Second, references to supply and demand shocks in oil markets are thrown around loosely these days — and sometimes incorrectly. For example, the decline in oil prices that began in 2014 is sometimes called a U.S.-based shale supply shock. But it is really the messy endgame of a demand shock that began in 2004, driven by rapid economic growth in emerging markets. The 2014-15 drop in oil prices began as the return of oil prices from a short-run, price-signaling level near $100 per barrel to a long-run equilibrium near $60.
The market’s recent overshoot to near $30 per barrel is a second phase of the ongoing price crash, and it results from events that weren’t foreseeable as the correction began in 2014. In particular, oil markets now are struggling with the return of Iranian exports and a global economic slowdown. The current price correction is brutal, coming immediately on the heels of the long-run adjustment to $60, and the pain is most acute for marginal suppliers like shale and oil sands. But wherever oil prices are today, they should be headed back to $60 in a matter of months.
Three basic tools
Let’s start with Figure 1 and three basic tools: the demand for oil, the short-run oil supply curve and the long-run supply curve. The per-barrel price of oil in today’s dollars is on the vertical axis, and the quantity supplied or demanded is on the horizontal. We have many studies about how these curves behave in oil markets.
- The curve DD represents the demand for oil, sloping down and to the right. The shape of the demand curve varies over time. It is quite inelastic (close to vertical) in the short-run when the stock of energy-using capital is fixed, meaning that oil consumption barely responds to price changes. In the long-run, as the housing stock is upgraded, new energy-efficient machinery is installed or fuel-efficient cars are produced, the curve flattens and we get a larger response of oil consumption to price changes. Two widely cited studies by Dahl and Cooper say that a 10 percent increase in oil prices results in only a 0.5% to 0.7% fall in short-run consumption, but a 2-3 % response in the long run. For simplicity, I use only a single curve DD, since the important distinction in this analysis is between short- and long-run supplies.
- The short-run marginal cost or supply curve is also nearly vertical, so a spike in the price of oil brings on little new production in the short-run. A series of statistical estimates by Krichene, for example, failed to find any short-run production response at all.
- The long-run adjustment of oil production to price appears to very long in oil markets. We might build a factory to expand widget production in 12 to 18 months. But the supply shocks of the 1970s persisted from 1973 until 1982, and the emerging market demand shock from 2004 to 2014. The stylized version of the curve in Figure 1 slopes up and to the right as new oil reserves come from higher-cost sources.
- Figure 2 is a more explicit representation of the long-run supply curve as we work to provide the world with 96 million barrels of oil per day. The first reserves developed are the least expensive, from the on-shore Middle East at $10-$25 per barrel, then the Offshore Shelf at $40, and then from a variety of sources that keep price near $50 until we need 85 million barrels per day. Then the price to bring on new supplies rises rapidly, with U.S. shale at $65, oil sands at $70 and Artic oil at $75. These marginal suppliers all find themselves on the cusp of the 96.3 million barrels produced in 2015. Looking back, it is hard to imagine the long-run price of oil slipping under $50. Looking forward, global growth in demand at 1.0 to 1.5 million barrels per year will require higher prices near $65-$70.
The 1970s Supply Shock
In the 1970s, U.S. oil production began a long decline, while domestic demand continued to grow. There were ample supplies of low-cost crude available from the Middle East, and the locus of power in world oil markets quickly shifted from the Gulf of Mexico to the Persian Gulf. One important result was the repeated involvement of the U.S. in the political turmoil of the region, beginning with the 1973 Arab Israeli War and the subsequent Arab oil embargo. Throughout the period, OPEC gained economic power and used it repeatedly to leverage political unrest into higher oil prices. By the middle of the 1970s oil prices (measured in today’s dollars) rose from $17 to $55 per barrel; following the Iranian Revolution in 1979, and the Iran-Iraq War in 1980, the price pushed well over $100.
Using our simple tools in Figure 3, suppose we are initially in equilibrium at price P1 and quantity produced of Q1. The OPEC cartel constrains production, and the short-run supply curve is shifted back from SR Supply1 to SR Supply2. A new short-run equilibrium is found at a much higher P2, and this high price becomes the signal to markets to increase global oil production.
In this case, the new price signal was seen and acted on, triggering a frantic search for oil reserves. In the 1970s, Alaska, the North Sea and Nigeria all large delivered new oil supplies that put downward pressure on oil prices. OPEC fought weakening prices, cutting production as prices slid from $100 to $60 per barrel between 1982 and 1986. OPEC’s share of global oil production declined from 49% to 28%, led by the Saudi decline from 15% to 6%.
In 1987, OPEC finally realized that – like King Canute – this rising tide of oil would never be reversed thorough their own efforts. To capture revenue, OPEC capitulated on price and began pumping at high levels; the price of oil collapsed to $20 in today’s dollars.
With the cartel resigning its position, shouldn’t we just return to the old equilibrium at P1, Q1? Unfortunately, no. By withholding its own reserves from the market for too long, and allowing the disequilibrium oil-price signal to stay too high, OPEC allowed production capacity to rise far above what was needed. Worldwide oil production in 1986 was 62 million barrels and the price was $60 and falling when OPEC suddenly added back 8.0 million barrels per day it had been holding off the market. World oil demand didn’t reach 70 million barrels until 1995. In Figure 3, shift the curve SR Supply3 to the right, resulting in price P3 that now is below long-run equilibrium.
Figure 2 tells us that the long-run equilibrium price for 62 to 72 million barrels per day of production should have been near $50 per barrel in current dollars. Instead, OPEC left a decade-long hangover at a below-equilibrium price that averaged only $35 from 1987 to 1995.
Emerging Market Demand Shock
The decline in oil prices that began in 2014 is sometimes described as a supply shock. It is true that from 2009 to 2015, oil production in the U.S. rose by 4.85 million barrels per day, accounting for all the increase in non-OPEC production. But I would argue that this new production was the response to a decade-old demand shock and the 2014-2015 price adjustment – at least initially – was the return to long-run equilibrium near $60 per barrel.
A supply shock in this context requires that the long-run marginal cost curve in Figure 2 shift down or to the right. In Figure 2, for example, if we found a new and unexpected source of 10 million barrels of $53 oil, it would displace all the higher cost sources on the right side of the curve – ultra-deep water, shale, oil sands, Artic drilling – and we would not need them until global demand reached well over 100 million barrels. It is hard to argue this was the story for U.S. shale. There was some technological innovation, but shale staked out its place as a marginal option at a relatively high price. The $100 price signal did much more to expand U.S. shale production than any innovation along the long-run marginal cost curve.
A better way to understand today’s decline in oil prices is as a response to the emerging market demand shock that began in 2004. Brazil, Russia, India and China accelerated growth, and as they raised their standards of living, they put upward pressure on the price of metals, food, agricultural raw materials and oil. The price of oil rose faster and further than other commodities, but prices all rose sharply. Figure 4 shows that since 2003, all of the increase in global oil demand has come from developing non-OECD countries, while demand from the developed nations was falling.
The textbook solution for a demand shock is shown in Figure 5. Demand shifts up from D1 to D2, and inelastic supply moves price up sharply from P1 to P2. Let’s say that this new price signal to expand capacity is P2 = $100 per barrel, and stays there for several years. We know this is well above the current long-term equilibrium price, which is $60 per barrels at 96.3 million barrels of global production.
I am simply suggesting that the current price correction than began at the end of the 2014 marked the end of the 2004 demand shock and required a shift in oil prices to a new $60 long-run equilibrium. In fact, that seems to be exactly what was playing out between April and June of last year. The price of oil stabilized near $60 per barrel, the domestic Baker Hughes rig count bottomed in June as drilling turned up briefly and oil-related layoffs came to a halt. It was as if the equilibrium price adjustment had fallen into place and a V-shaped drilling recovery was underway. This was no replay of 1987 with an OPEC build-up of surplus capacity – it now holds less than 2 million barrels per day – and no decade-long wait to work off that surplus.
Oil Prices at $30 and below?
How do we find oil at $30 per barrel today? Two subsequent events turned last summer’s fragile equilibrium into a rout: the Iran Nuclear Agreement signed in July 2015 and the devaluation of Chinese currency that followed in August. Economic sanctions against Iran were imposed in 2011, and Iran’s daily oil exports fell by about 1.2 million barrels. With the lifting of sanctions, Iran has made it clear that it plans to quickly regain its previous export position. Our long-run supply curve just got a million barrels per day longer, starting with low-cost onshore Middle East production that must be absorbed before high-cost shale oil returns.
Meanwhile, China is making a tricky transition from a manufacturing and export-led economy to a consumer-driven economy, and it has long been anticipated that annual Chinese economic growth would slow to near 6%. Concerns are frequently expressed that China might not make the legal, financial, labor market, energy and other reforms necessary to continue even on this more modest growth path. But the August devaluation of the yuan, accompanied by significant turmoil in Chinese stock markets, distilled these concerns into real fear about the Chinese economy – fear that quickly spilled into oil markets.
I leave it to you to guess if we should shift the oil demand curve down – and how far we lower the short-run equilibrium price. The forecast for 2016 from the International Energy Agency in Table 1 more or less dismisses the China growth problem, seeing India filling any demand void left by China. IEA points to slow growth in Europe and Latin America as the source of poor demand growth in the global economy.
That said, unless there is much more to the China story than is now apparent, this should be a routine correction for oil markets. There is no massive 1987 supply shock and the surplus currently driving oil prices should be resolved in a matter of months. How many months before oil returns to a long-run $60 or $65 per barrel? 6 months? 12? 18? That is the difficult and painful detail that remains to be resolved.
 Noureddine Krichene, “A Simultaneous Equations Model for World Crude Oil and natural Gas Markets,” IMF Working Paper WP/05/32, February 2005.
 Figure is adapted from a chart that seems to be based on Morgan Stanley research, but is widely copied in many places on the internet. For example, http://www.eclectecon.net/2014/12/marginal-costs-of-oil-production.html
Bill Gilmer is director of the Institute for Regional Forecasting at the University of Houston’s Bauer College of Business. The Institute monitors the Houston and Gulf Coast business cycle, analyzing how oil markets, the national economy and global expansion influence the regional economy.