This note was originally published at 8am on June 21, 2012. INVESTOR and RISK MANAGER SUBSCRIBERS have access to the EARLY LOOK (published by 8am every trading day) and PORTFOLIO IDEAS in real-time.
"Our knowledge of the factors which will govern the yield of an investment some years hence is usually very slight and often negligible."
- John Maynard Keynes
That’s a quote from Chapter 12 of Keynes’ General Theory, “The State of Long-Term Expectation.” Much of Keynes’ work was marginalized by the economists that were influential shortly after him – namely John Hicks – particularly the existence and importance of uncertainty in investment. As a result, what is today considered Keyensian Economics is hardly the economics of Keynes (Steve Keen, 2011). Hyman Minsky – the preeminent economist on fragility in financial markets – wrote in 1975 that, “Keynes without uncertainty is something like Hamlet without the Prince.”
This Early Look is not on Keynes – you’re welcome – but on uncertainty and oil prices.
As an energy analyst, I read a lot about oil markets. One of the latest topics #trending on the subject is the price of oil approaching the marginal cost (MC) of production; this one always seems to get popular when oil prices drop precipitously, as it’s a go-to reason to ‘buy-de-dip,’ say the perma-bulls.
The MC of production is the change in total cost that comes from producing one additional unit of a good – a bicycle, a bushel of corn, a barrel of oil. It is an important economic concept that determines the price of that good for a given level of demand. On a long enough duration, the price of oil will converge around the MC because should price fall below the MC, the MC producers will not put incremental capital to work to arrest natural production declines, leading to a shortage and rising prices; and should price rise above the MC, it incentivizes production above what is demanded, leading to a surplus and falling prices.
Yes, MC matters, and yes, oil prices will track it over the long-term. But the idea that the MC of oil production, and with it the oil price, is permanently headed higher – a popular opinion – is a fallacy. Consensus is comfortable, and after a decade-long rise in the MC from $25/bbl in 2001 to $90/bbl in 2011 why would the next ten years be any different? But the future is, of course, unknowable, and to deal with that inherent uncertainty “the facts of the existing situation enter, in a sense disproportionately, into the formation of our long-term expectations” (Keynes, 1937). In other words, only after oil prices have increased 15% p.a. since 2001 does a deflating oil price seem implausible.
In general, though, commodities – even non-renewables (fossil fuels) – do not tend to hold their real value over the long-term, proving poor investments. New technologies, greater efficiency in extraction and production, and the substitution of one commodity for another (at one time our primary fuel was wood, then it was coal, today it is oil) drive the MC of production lower, and with it real prices. After adjusting for inflation, major industrial commodity prices fell 80% between 1845 and 2002, though regained some of that lost ground over the last decade in a fantastic commodities bull market (The Economist, 2011). Only gold and oil have held their value in real terms, and, indeed, oil has been a spectacular investment over the last decade, gaining 300%. It is easy to forget, though, that the real price of oil in 2000 was no higher than it was in 1950.
We can point to a number of reasons why the MC of oil supply and the nominal price of oil have meaningfully outpaced inflation over the last decade, though easy money and China’s incredible investment boom top our list. Others point to the decline in easily accessible oil reserves and rising social costs of Middle East nations, but we see less validity in those theses. New technologies and greater efficiencies can counter harder-to-reach reserves; and oil exporters’ government budgets are pro-cyclical.
In fact, most costs are pro-cyclical. The relationship between the MC of oil production and the oil price is a classic example of mutual causality. Is the price of oil higher because rig rates, steel pipe prices, production taxes, and labor costs are higher? Or is it that rig rates, steel pipe prices, taxes, and wages are increasing because the oil price is? Both occur simultaneously and as a result, the MC of oil production is just as much a moving target as the price is – not some Rock of Gibraltar level of support.
We’ve seen this movie before. In constant dollars (year 2000) the price of US coal decreased from $31.40/short ton to $16.84/short ton between 1950 and 2003; in other words, after adjusting for inflation, coal prices have fallen more than 1% p.a. since 1950. This trend is due to a decline in the MC, or “marked shifts in coal production to regions with high levels of productivity, the exit of less productive mines, and productivity improvements in each region resulting from improved technology, better planning and management, and improved labor relations” (Edward J. Flynn, 2000). Those trends persist today, and the high cost coal producers are gasping for air (pull up a 5Y chart of PCX or JRCC).
And of course, we have US natural gas, which has fallen in nominal terms from $12/Mcf in 2008 to $2.50/Mcf today in a stunning display of how technology and innovation can lower the marginal cost and price of a fossil fuel. Visions of permanently high natural gas prices prompted a build-out of LNG import facilities in the mid 2000’s. Alan Greenspan (clearly an expert on the subject!) said in 2003 that, “high gas prices projected in the American distant futures market have made us a potential very large importer.” With impeccable timing, Cheniere’s Sabine Pass received its first cargo of imported gas in April 2008, and now that same visionary company will spend $6.5B to export gas by 2017 (nat gas bottom?). And this wasn’t the first time the gas bulls were fooled. After years of rising natural gas prices in the 1970’s, four major LNG receiving terminals were constructed only to see gas prices peak in 1983 and decline for the next 15 years; three of those plants were eventually mothballed.
Is the rapid increase of the MC of oil production and oil prices any more “secular” than it was for natural gas in the 2000’s? Will Chinese demand for commodities grow at the same pace over the next ten years as it did for the last ten? What would sustained US dollar strength do to commodity prices, oil in particular?
The confidence one can have in answering such questions is limited, perhaps even “negligible” (Keynes), but few in this Game accept that. It’s funny – you don’t often hear investors or analysts say, “I don’t know,” but when consensus is proven really wrong, the all-too-common excuse is, “How could I have seen this coming?”
Our immediate-term support and resistance ranges for Gold, Oil (Brent), US Dollar Index, EUR/USD, and the SP500 are now $1590-1614, $91.20-96.79, $81.21-82.02, $1.24-1.27, and 1333-1362, respectively.