why have oil production costs fallen so much?

rules for commodities

From years of analyzing oil, gas and metals mining–as well as watching agricultural commodities and high-rise real estate out of the corner of my eye–I’ve come to believe in two hard and fast rules:

–when prices begin to fall, they continue to do so until a significant amount of productive capacity becomes uneconomic and is shut down.  That’s when the selling price of output won’t cover the cash cost of production.  Even then, management often doesn’t reach for the shutoff valve immediately.  It may hope that some external force, like a big competitor shutting down, will intervene (a miracle, in other words) to improve the situation.  Nevertheless, what makes a commodity a commodity is that the selling price is determined by the cost of production.

–it’s the nature of commodities to go through boom and bust cycles, with periods of shortage/rising prices followed by over-investment that generates overcapacity/falling prices.  The length of the cycle is a function of the cost of economically viable new capacity.  If that means the the price of new seed that sprouts into salable goods in  less than a year, the cycle will be short.  If it’s $5 billion to develop a gigantic deep-water offshore hydrocarbon deposit that will last for 30 years, the cycle will be long.

boom and bust spending behavior

During a period of rising prices, cost control typically goes out the window for commodity producers.  Their total focus is on adding capacity to satisfy what appears at that moment to be insatiable demand.  Maybe this isn’t as short-sighted as it appears (a topic for another day).  But if oil is selling for, say $100 a barrel, it’s more important to pay double or triple the normal rate for drilling rigs or mud or new workers–even if that raises your out-of-pocket costs from $40 to, say, $60 a barrel lifted out of the ground.  Every barrel you don’t lift is an opportunity loss of at least $40.

When prices begin to fall, however, industry behavior toward costs shifts radically.  In the case of oil and gas, some of this is involuntary.  Declining profits can trigger loan covenants that require a firm to cease spending on new exploration and devote most or all cash flow to repaying debt instead.

In addition, though, at $50 a barrel, it makes sense for management to:  haggle with oilfield services suppliers;  do more ( or, for some firms initiate) planning of well locations, using readily available software, to optimize the flow of oil to the surface;  optimize fracking techniques, again to maintain the highest flow; streamline the workforce if needed.   From what I’ve read about the recent oil boom, during the period of ultra-high prices none of this was done.  Hard as it may be to believe, getting better pricing for services and operating more efficiently have trimmed lifting expenses by at least a third–and cut them in half for some–for independent wildcatters in the US.


This experience is very similar to what happened in the long-distance fiber optic cable business worldwide during the turn of the century internet boom.  As the stock market bubble burst and cheap capital to build more fiber optic networks dried up, companies found their engineers had built in incredibly high levels of redundancy into networks (meaning the cables could in practice carry way more traffic than management thought) and had also bought way to much of the highest-cost transmission equipment.  At the same time, advances in wave division multiplexing meant that each optic fiber in the cable could carry not only one transmission but 4, or 8, or 64, or 256…  The result was a swing from perceived shortage of capacity to a decade-long cable glut.

My bottom line for oil:  $40 – $60 a barrel prices are here to stay.  If they break out of that band, the much more likely direction is down.


which recovers first, crude oil or base metals?

I’m in the oil first camp.  (My private opinion is that it could take a decade or more for base metal prices to perk up.  Whether that turns out to be true or not is less important to a long-only investor like me than the idea that recovery is not soon.)

How so?


Leading with (the opposite of what you’re supposed to do) my weakest reason, look at the last cycle of gigantic investment in expanding natural resources production capacity.  Oil and metals prices both peaked in 1980-81  …and then plunged.  Oil stabilized and began to move up again in 1986; for metals recovery was over a decade later.

closing the supply/demand gap

There’s only a gap of a couple of percentage points between the amount of oil the world is demanding and the amount producers are willing to supply.  Growth in the car industry in China, the replacement of scooters and motorcycles with cars in other high-population countries like India and the strong increase in gasoline consumption in the US now that prices are lower all argue that the shortfall between demand and supply is, little by little, being erased.

On the other hand, the extent of base metals overcapacity is less easy to put your finger on, but is, nevertheless, massive.  Demand is also more cyclical–therefore less dependably growing, as well, but that’s less important than that mining capacity is added in gigantic chunks.

the nature of the enterprise

The up-front cost of a base metals mining project is very high.  There’s the mine itself, the huge machines that rip the ore out of the earth and the sometimes elaborate plants that crush or grind or otherwise separate it from the ordinary dirt.  Then there’s the transport link with the outside world.  All of this infrastructure can lie fallow for long periods without impairing the mine’s ability to be restarted–even expanded from its prior size–very quickly.

For oil, in contrast, finding new fields is a much more important issue.  Drilling new wells in an existing field is, too, in many cases.  As time has passed, the focus of the big oil majors has increasingly been on mega-projects that make them look much more like base metals miners than they did when I was covering the oil industry as a securities analyst in the late 1970s – early 1980s.

Hydraulic fracturing, however, has changed the industry for good.  This technology has made huge numbers of projects economically viable that have limited output that goes on for relatively short periods.  This converts 21st century oil exploration, in the US at least, into a sharp-pencil engineering business that even small firms can excel at.  Granted, the fact that production can turn on very rapidly when prices are high enough puts a cap on how far they can rise.  But the fact that several millions of barrels of daily output can be turned off equally quickly argues that the response time of the oil industry to a supply/demand imbalance will be much quicker than has been the case in the past.


oil: will falling prices reduce supply?

Ultimately, yes   …but only at lower prices than today’s., I think.

With any mining commodity, price declines normally end only when the highest-cost firms have to pay more to produce the commodity than they can sell it for.   Even then, if a production process is hard to restart or if the producers fear losing skilled workers permanently if they shut down, production often continues for a period even though cash flow is negative.

Petroleum has been an exception to this rule.  Oil had a period in the early 1980s when Saudi Arabia reduced its oil production dramatically in a vain bid to stabilize prices.  But its efforts were undone by other members of OPEC who agreed to cut production, too, but upped it instead to fill the vpoid left by Saudi cutbacks.  It took Saudi resumption of production and a consequent plunge in prices for the others to fall into line.  This time around Saudi Arabia has made it clear it won’t repeat its production-cutting mistake.

If cartel action won’t stop the current oil price decline, then we’re left with normal commodity forces to do the job.  The most likely production to shut down for cost reasons is output generated through hydraulic fracturing in North Dakota and Texas.  Estimates of cash production costs for fracked wells ranges from $40 – $60 a barrel.  In theory, therefore, production won’t be taken off the market until prices reach $60.  Even at that level, however, only a small amount of output will probably be lost–not enough for price stabilization.

One wild card:  bank loans.  Typically, smaller oil exploration companies of the type that have been successful with fracking try to boost their returns or speed their expansion by leveraging themselves financially.   Except in times of speculative excess, bank loans.to exploration companies contain restrictive covenants.  These normally mandate that the explorer must maintain reserves with a value of, say, 3x the amount of the loan.  If the value of reserves falls below a certain minimum, say 2x the value of the loan, the borrower is required to devote most or all of its cash flow to repaying its borrowings.  In other words, it can no longer pay a dividend to shareholders nor can it spend money on new drilling.  This last is potentially a big issue for frackers, whose wells tend to have relatively short productive lives.

My guess is that borrowings of the type I’ve just described will ultimately be the reason the oil price ultimately stabilizes, by halting the growth of fracking.  Two ways to gauge whether this is happening:  dividend cuts, and reductions in the number of new wells started.