The Seventh Fold

Last week, I alerted you all to some interesting preliminary findings of my ongoing research into oil exports.  The news that world oil exports peaked in 2006 probably came as a surprise, and a troubling one at that.  The post this week is no less troublesome.  Through a peak-oil-export enlightened analysis of diesel crack spreads and diesel exports data, I show that during the heat of the price run, US diesel consumers (i.e. our transportation sector) were for the first time in history outbid by foreign consumers.

Here in the U.S., roughly 80% of diesel is used for goods movement, and none of the existing fleets of trucks and trains can can be fueled by anything but diesel.  While the U.S. is a trade dependent nation, individual states are even more so.  With $27 million of freight in transit on Washington roadways every hour of every day, and with 46% of jobs in freight dependent industries, Washington is the nation’s most trade dependent state (link)  (.pdf warning).

No. 2 Distillate (diesel) Fuel Consumption by End Use. Source: Jim Hansen

Consequently, the Washington economy is especially vulnerable to diesel price surges.  In January of 2007, the price for diesel was already $2.50 – a multi-decadal high.  Then, over the next twelve months, the price for diesel doubled, and the rising price of crude oil only explains 75% of the increase in price of diesel.  What explains the 25% difference?

The answer is hidden in two statistics: the diesel crack spread and diesel exports.

Crack spreads measure the difference between the value of a barrel (42 gallons) of crude oil and a barrel of refined product.  If the price for a barrel of diesel is $105 ($2.50 per gallon), and the price for a barrel of oil is $100, the crack spread is $5.  Historically, the crack spread for gasoline is roughly equal to the crack spread for diesel.  The reason for this is that refiners are usually pretty good at estimating end-use demand, and they have some leeway in adjusting the refining process to maximize the output of the most profitable refined product.

The first of the following graphs (Figure 1) shows the crack spreads for gasoline (red) and diesel (blue).  The second graph (Figure 2) shows the difference between the price (per barrel) of diesel and gasoline.  Both of these figures show that in the decade before the price run, crack spreads fluctuated very little from the $28 to $32 band, and that gasoline was usually the more expensive of the two, though the difference rarely exceeded $5 per barrel.

Figure 1: Diesel and Gasoline Crack Spreads: 1994 – 2009 (source: theseventhfold.com and The US Energy Information Administration)

Figure 2: Difference in the Price of Diesel and Gasoline: 1994 – 2009 (source: theseventhfold.com and The Energy Information Administration)

In September of 2004, this trend promptly reversed (see Figure 2).  Rather suddenly, the market price for diesel overtook the price for gasoline.  Then in mid-July 2007, the crack spread went, in a word, “crazy”.  The crack spread for diesel skyrocketed from $60 per barrel to $70 per barrel, and ultimately to almost $80 per barrel in August of 2008.

Why did these historical trends suddenly flip-flop?  And why did the diesel crack spread so suddenly gap the gasoline crack spread?  And why weren’t refineries able to predict this reversal and adjust refinery output to take advantage of diesel’s favorable crack spread?

Take a look at what happened with diesel exports over this same period of time (Figure 3).  Between March 1994 and July of 2006, diesel exports averaged roughly 5 million barrels per month (~170,000 bpd).  By August 2008, the amount of diesel exported grew by more than 500%.

Figure 3: US Exports of Diesel: 1994 – 2009 (source: theseventhfold.com and The Energy Information Administration)

Turning to US consumption (pink line in Figure 4, below), we see two clear periods emerge from the data.  Between 1994 and November/December 2007, diesel consumption trended upward (though seasonal variation causes predictable, cyclic volatility).  Following the economic collapse that officially started in December 2007 (though I would argue that the great unwind really began in 2006), diesel demand began a drastic slide bottoming out in August 2009.  We do see that diesel demand began climbing again in October, but much of this gain is cyclical.  Only time will tell, but it looks to me like this uptick is not a ‘break-out’.

Figure 4: Domestic Diesel Consumption and Exports: 1994 – 2009 (source: theseventhfold.com and The Energy Information Administration)

So what does this all mean?  In my previous post (link), I released data showing that global crude oil exports peaked in 2006.  Consequently, this peak coincides with a very steep climb in oil prices.  As the price of oil rose, so too did the price of diesel.  But the US was not the only country to suffer from the constriction of net exports.  ALL net importers, in fact suffered.

Bear in mind that there is relatively very little international demand for gasoline – at least when compared to diesel.  So, as oil exports peaked, so too did the amount of diesel that could be produced worldwide.  This led to diesel shortages not only in the US but in the pacific basin more generally (thanks to Jim Hansen of Ravenna Capital Management for pointing this out).

In turn, increased competition for diesel drove both the price and the crack spread for diesel way up.  What was happening, was that diesel prices needed to rise enough bid the marginal consumer out of the market.

What is surprising and troublesome is that the export and crack spread data show that US diesel consumers were – for the first time ever – outbid by foreign consumers! This marks a watershed moment in US history.  Think about it, trade forms the backbone of the national economy, and we use diesel to move goods.  When US consumers (read: freight carriers) were priced out of the market, goods movement and trade was forced to slow and economic output was constrained.

But the linkages and impacts don’t stop there.  Turning to Washington state,  

The Tesoro Refinery... before the explosion which killed six.

we see that during the heat of the diesel price surge, the rising crack spread made it economically viable for the Tesoro refinery in Anacortes, WA to upgrade their facility to crack more of the heavier refined hydrocarbon products – namely asphalt oil – into diesel.  As my friend Jim Hansen points out, this shift in refinery output led to an asphalt shortage in the state which caused some road projects to be delayed.  Jim even sent a great link for further reading on the subject.

Let me wrap up this post with a quick note on where Washington is situated today.  Currently, the price of diesel in Washington is slightly higher than the price of gasoline.  In part this is due to the continued higher than average exports of diesel, but in part this is due to the recent explosion at the same Tesoro refinery.  Unfortunately, this explosion killed six workers.  My thoughts and condolences go out to their families.

My take is that between the peak in global oil exports, and the fact that foreign competition for domestically refined diesel is now able to outbid US consumers, any *real* (read: non-financial) economic recovery will be greatly constrained by the emerging energetic reality.  We need to restructure the economy around liquid fuel constraints or wave goodbye to times of plenty.