In a previous post I made my case that, in order to bring Newt Gingrich's campaign promise to bring gasoline prices to "$2.50/ gallon or below," the U.S. would need to bring enough crude production to market to bring the price per barrel (for Brent crude; WTI, or West Texas Intermediate is the figure most often quoted on the radio and typically is quite a few dollars lower) of around $68.00. The spot price of Brent crude closed Friday, April 20 at $118.52/bbl (as contrasted with WTI at $103.88/bbl).
Can the U.S., with all regulatory hurdles removed, supply sufficient oil to the market to bring the price down by 42%? And would the extraction be economically viable at a price of $68.00/bbl? Let's take a look.
This will only be from a 30,000 foot level - there are many subtleties involved in U.S. gasoline prices, else how to explain the huge difference in price per gallon between Tulsa, OK and Chicago, IL ($3.459 vs. $4.464)? But crude oil is sold into a worldwide market. If there's a significant discrepancy (with quality taken into account) between the price that willing buyers will pay in disparate geographic regions, arbitrageurs will quickly step in to profit from the difference and equilibrate the prices.
This is one of the purposes of the various Commodities Exchanges (the others are things like the ability to hedge and to speculate) and they provide a very efficient market. The discrepancies above can be attributed to proximity to refiners, proximity to pipelines, specific blending requirements and taxes of states, etc. It wouldn't be possible to buy 20,000 gallons in Tulsa for $69,180, truck them to Chicago and sell them for $89,280 and pocket the $18,800 difference (after paying about $1,300 for the trucking).
In any case, the price of oil is demonstrably very sensitive to small changes in either demand or supply. An economist might say that price elasticity of both supply and demand are, at least in the short term, very low. That is, the graphs that show price on the vertical axis and quantity demanded or supplied on the horizontal axis are close to vertical. In an efficient market, the point where these lines cross sets the so-called "clearing price," that is, where the price is that which causes supply and demand to be equal.
What we want is the price elasticity of supply (PES) so that we can figure what amount of additional supply would cause the price to decrease from $118/bbl to $68/bbl. This is, of course, a simplistic analysis in that it assumes assumes that "all else is equal." Among other things, this assumes that demand doesn't rise as prices drop (though, since the demand curve is inelastic, this might not be a bad approximation in the short run), that OPEC and non-OPEC suppliers don't react by lowering production (they might not since they count on the foreign exchange income), and other considerations.
PES is defined as % change in Quantity supplied divided by % change in price. Estimates range from 0.1 to 0.01. Using a baseline of 85 million bbl/day and plugging in the numbers represented by a decrease in price from $118/bbl to $68/bbl, it can be calculated that the increase in production needed would be about 3.6 million bbl/day at an elasticity of 0.1. At 0.01, it would be a mere 600 thousand bbl/day.
The second number is highly unlikely to be correct in that it was calculated by using production figures as prices rose at a time when most producers were unable to increase production (regardless of their protestations to the contrary). There are a few more seemingly authoritative sources that use the 0.1 figure and this excellent essay discusses the interaction between supply and demand in recent history and suggests (through implication) that the interaction leads to an effective doubling of elasticity.
In any event, the sharp eyed will have noticed that the implication here is that a price decrease would result in an increase in quantity supplied. This is, of course, not what's implied in a discussion of supply elasticity. But the supply elasticity is so close to vertical that what we're really looking for is the rightward shift of the quantity supplied curve that, in combination with the quantity demanded curve, would clear the market at $68/bbl. I simply used the negative of the PES in the calculation.
This very informal (and, no doubt, flawed) analysis indicates that an increase in world production on the order of between 3 and 4 million bbl/day, ceteris parabus, could cause a decrease in crude prices sufficient to drive the cost of gasoline to $2.50/gallon or less.
Next I'll discuss whether the U.S. technically recoverable reserves and the available equipment would allow such an increase.
Sunday, April 22, 2012
On Intrade.com, as far as I know the best and most extensively used prediction market, it's estimated that President Obama's chance of being re-elected in November, 2012 are at about 60% and the chances of the Republican Party adding control of the Senate to its existing control of the House of Representatives are about 65%.
Say what you will about prediction markets' accuracy in comparison to other forms of prognostication, they've typically exhibited accuracy at least as good as poll results (though the markets are heavily influenced by poll results) and it's awfully easy to get moment by moment snapshots as trading takes place.
Of course, the voting results for President and for the Senate are far from independent but, if they were and if the chances above could actually be treated as probabilities rather than betting odds (see below*) then there's about a 40% likelihood that we'll have a Democratic President and that both houses of Congress will have Republican majorities.
This is not atypical of the second term of a President, but with the level of toxicity exhibited by present-day political discourse and a significant degree of public discontent (cf. the Occupy Movement and the Tea Party Movement) this seems to me to be a recipe for increasing levels of public disorder and chaos.
And, on a subject near and dear to my heart and that which is the foundational area of concentration of this blog, this is not a political environment in which the energy Sword of Damocles hanging over all of our heads will be able to be blunted.
*The salespeople in my Company must enter a "probability of win" when they log a proposal into our CRM database. We all know what we mean by this but, logically, it makes no sense. It's a one-time event and the probability is either 0 or 1, and we'll know what it was when the results are in. It's not as if we can say "in 100 similar opportunities that we've pursued in precisely this way in precisely similar environments we have won 60 of them." In order to reconcile this to the logical framework within which my thought processes operate, when a salesperson enters 60%, I interpret this to mean that he or she would take either side of a 3 to 2 bet where the bettor against us was offerred 3 units if we fail, and we win 2 units if we succeed.