Another picture that tells a story

This is the current picture of the US gasoline stocks, as posted at the EIA site.  You need to scroll down the site to see the graphs. The recent fall is quite marked.  But I would like to reiterate something that Edward Tufte has taught me (by reading his books), and that is that you have to recognize that this is only highlighting the shifts.  You have also to look at the scale on the left to see that while the drop has been about 30 million barrels, we still have about 190 mb left, and so there is no need to panic.  

Normally, as you can see, we would be entering a period where we would be rebuilding stocks.  The damage from Katrina and Rita, and the current refinery problems, will likely see that delayed.  Trying to balance gas production against heating oil may be a hard call that someone is going to have to make.

I'm staring at that graph and the blue average band doesn't look right. The left hand side and the right hand side should wrap into each other. The values match, but there's a completely sharp change in first derivative at both the top and bottom of the blue band. I can't think of any plausible explanation for that.
It looks like sloppy graph drawing to me. I think the data is based on monthly increments, and the graph drawer made a notional effort to do some curve fitting, but didn't work very hard at it.
Also don't forget that the lines plotted over these 'average bands' are themselves averages. The weekly EIA reports use 4 week averages to report all numbers.

Averages can hide things... when traders react to X being over or under the 5 year average, its often without benefit of realizing that the "average band" would be higher if a smaller averaging period were used. 2001 was an anomoly year... and demand has been very strong over the past 2 years:

1.2% 2000 (for comparison)
0.7% 2001
1.0% 2002
1.6% 2003
3.1% 2004
3.8% (2005 est, it will be lower, but still high)

Stocks must be near the high end of the 5 year range or we are in trouble!

MMS Status - oil shut out edged up a little higher from 1,527,630 BOPD. NG shut in down ever so slightly. Now 5 days in, almost virtual GOM total shut down.

Today's shut-in oil production is 1,512,937 BOPD.  This shut-in oil production is equivalent to 100% of the daily oil production in the GOM, which is currently approximately 1.5 million BOPD.

Today's shut-in gas production is 7.856 BCFPD.  This shut-in gas production is equivalent to 78.56% of the daily gas production in the GOM, which is currently approximately 10 BCFPD.

The cumulative shut-in oil production for the period 8/26/05-9/27/05 is 36,361,383 bbls, which is equivalent to 6.641 % of the yearly production of oil in the GOM (approximately 547.5 million barrels).

The cumulative shut-in gas production 8/26/05-9/27/05 is 172.506 BCF, which is equivalent to 4.726% of the yearly production of gas in the GOM (approximately 3.65 TCF).

It's the 5 year average - if there was a big net annual draw or build in the first or last of the 5 years, the figure would not necessarily be periodic.

BTW - I don't buy the "30 million down, but 190 million left so no problem" argument.  There needs to be enough gasoline in the system to keep things flowing - think of natural gas, where there has to be enough gas in storage to maintain pipeline pressure.  No one seems to publicize the number at which the supply system breaks down...

Trying to balance gas production against heating oil may be a hard call that someone is going to have to make.
Given the tendency by the current administration to prefer (or at least give lots of lip service to) free markets over government intervention, I don't see anybody making a hard call about balancing gasoline vs. heating oil production except for Adam Smith's invisible hand. Unfortunately, I think that this may lead to a situation where refineries respond to the economic incentives and don't produce enough heating oil, and thus poor people freeze to death because rich people won't stop driving their SUVs. The poor don't have much standing in the free market.
But the poor do, in effect, have a standing in the House and the Senate, where lots of Republicans are eager to get re-elected in 2006.  If things got bad and Bush sat on his hands, the Legislative branch would take action and, if needed, pass something over a Bush veto.  It would be political suicide not to.  (Although in the interest of fairness, I have to point out that the $50B appropriation for New Orleans passed with 11 Republican House members voting against it.)

There are already Republicans who are distancing themselves from the Iraq war, including the guy who started the "Freedom Fries" idiocy.  A domestic heating emergency would have Republicans jumping ship in droves.

Now, whether the Congresscritters would actually do the right thing is another, and legitimate, question entirely.

More than likely, it would indeed by the Congressman paying lip service to their constituents.  Bundled inside of any $10 billion dollar social program bill would be $20 billion in business welfare to "help encourage new growth."  

I think people need to stop and remember that neo-liberal laissez-faire is only one directional, and that our representatives only work for us enough to pretend like they respect our intelligence.

Historically, poor people don't vote.  Or should I say not many of them vote.
Sealth: Refineries respond to the economic incentives every year and shift production to distillates for heating oil during the fall. This year, they will do the same thing. The only difference is that the prices will be higher for both heating oil and gasoline. Demand for each determines the level of production and price for each. There is no reason to think that heating oil is less profitable than gasoline.
There is no reason to think that heating oil is less profitable than gasoline.  Yep, he's right - for instance at 12:19 today, NYMEX

Jan Heating Oil - 2.1160

Jan Gasoline - 1.9500

seems to be some money in Heating Oil....

Yes, but if prices are high enough that demand destruction (a euphemism for "the poor can do without") is occuring in both gasoline and heating oil (a distinct possibility this year), then not enough heating oil will be produced to allow everybody to adequately heat their homes; thus, the rich will continue to drive SUVs while the poor freeze to death.

In another part of this thread, somebody pointed out that Congress could dictate to refineries that they have to produce enough heating oil to supply all residential heating requirements for the winter, even if it cuts into their bottom line (because the rich are willing to pay more for gasoline than the poor are able to pay for heating oil). My question is whether or not Congress will be able to muster the political will to do so (after fending off accusations of being communists) in time to really make a difference this winter.

This all may be moot because we are not currently short, and with any luck (and perhaps a mild winter) we can avoid a heating crisis this year.

If heating oil prices stay anywhere near the level cited by Fatbear, refineries will shift as much of their production as possible to take advantage of the it. Heatng oil prices higher than gasoline mean great margins for refiners. However, their equipment will be the main constraint on how far they can go.  With a given set of refining equipment, there is a fixed range of gasoline/distillate production levels.  The government can't order refineries to produce more than they can. However, you are right that the poor may not be able to bear the increased costs created by a product shortfall.
There's another thing that's even weirder. I measured carefully to the bottom of the sharp V in the red weekly line. I make that about September 2nd, give or take a day or two. So the picture appears to say that from July up until the point when Hurricane Katrina hit, stocks were dropping faster than usual (going from the very top of the average range to the very bottom). Then, right around the time when Katrina hit and wasted a bunch of refineries, stocks started going up faster than usual (reaching around 25% of the way up the band). This makes no sense!

Well, I guess there is one logical possibility: Katrina's effects on demand were larger than her effects on supply. That's a pretty considerable effect!

I guess this actually makes a lot of sense looked at under an efficient market hypothesis. Stocks have reached the bottom of the band and then a bunch of refineries get wiped out. The market, well aware that stocks are low, freaks and the price goes through the roof. The market gets subsequent stock reports and sees that the price is high enough to start to finally have some impact on consumption; the price starts to drift down.
Another possibility is that consumption didn't respond much to price but did respond to the fact that a million or so (very rough guesstimate) vehicles were suddenly taken off the road, and this being the deep South, that they were on average large fuel hogs, even for our country which is addicted to fuel hogs. If the average vehicle gets 20 mpg, is driven 300 miles per week, and their are a million of them out of service, that is a demand reduction of (1 million)*(300/20)/42=almost a third of a million barrels per week.
Well, but they aren't off the road, are they? They're driving back and forth between NO and Baton Rouge...
So which are there more of? The ones submerged in New Orleans or the ones driving back and forth into the disaster zone. Additionally, upwards of half a million people are no longer commuting because they have no job.

Another question is how we measure the inventory levels. I doubt that gas in one's gas tank is counted, so the sudden dip and rebound could be explained by everybody panicing and topping off the week before Katrina hit and then not buying the following week. This is, admittedly, a short term effect, but it might explain some of this.

FWIW, I am not looking at inventory or production or consumption numbers too closely at the moment because I think that the data is suspect (at best) due to the chaos on the Gulf Coast. I rather expect the numbers will all make a lot more sense around November (barring any further shocks or chaos, of course ... my own personal force majeure clause).

Bingo! Gasoline demand is price inelastic in the short run, but it's not perfectly inelastic.

(The definition of elasticity is based on the ratio of the percentage change in demand to the percentage change in price. A ratio greater than one is elastic, a ratio less than one is inelastic. A ratio of zero (price changes but demand doesn't budge) is perfect inelasticity.)

Econbrowser had some relevant discussion a little while back. Seems to me like there might be certain psychological points where a price increase has more impact on demand than others. If it's creeping up and creeping up, people complain a little bit but don't really take much action other than not spending the money on other things that they had to spend on gasoline. But then, when it hits $3, it causes people to sit up and go "Holy s**t", and really make some change that has a significant effect on their usage.
Although, what goes into determining the gasoline stocks? What if release of the SPR is only factored into the gasoline stocks after Katrina? Wouldn't that also account for an increase in stocks?
The latest increase in inventories occurs during the IEA emergency release of 1Mbbl/day of gasoline. Coincidence?
Wouldn't that only just be starting to arrive now though? They needed to charter tankers and then ship it across the Atlantic.
Ahh, I've got it - finally!

The inventory is the stocks in the supply chain.  As soon as you've bought gasoline, even if it's stuck in a storage tank in Holland with no tanker due for weeks, it belongs to you and you can include it in your inventory reports to the IEA.

Finally, an explanation of how Gasoline stocks can shoot up in the absence of 5% of national refining.

A better drawn version of the chart can be found Here (PDF)

Incidently, if you look Here (PDF) which is the same thing for January 2004, you can see they used to include a 'Lower Operational Inventory' line - it dissapeared some months ago.


EIA, not IEA of course
Ok, so if we were to remove the emergency stock release numbers from the approximately 8Mbbls EIA stock increase during the last two weeks we would end up with a net total stock loss of 6 (14 - 8) Mbbls.

Would this make sense?

Interesting. Is that operational inventory line supposed to be an estimate of how low things have to get before there are shortages somewhere in the system?
Well, I take it back - read the notes on the report, and stock reports are supposed to only include landed gasoline etc.

So I'm back to: the numbers don't seem to add up...


I think it's more interesting to look at the stocks in term of equivalent number of days of consumption (stocks / gas demand):


If the EIA numbers can be trusted, the US lost only one day of consumption after Katrina.

nice work.
you have to be careful. the EIA numbers are mostly measure of primary storage. this excludes both secondary (gas stations) and tertiary (your gas tank) storage. So if there has been flucuations in either of those, which seems likely of late, then one should be careful not to overinterpret week over week changes in the EIA numbers, especially in the gasoline demand number itself. If gas stations were providing consumer demand but not restocking then the demand number the EIA puts out would be depressed.

Someone said its good to look at the 'days' of inventory left vs. the absolute level of gasoline stocks. A standard measure of this is inventory level over an average of coming demand, or 'forward cover'. Before hurricanes US gasoline stocks were reaching for an all time low in forward cover, something under 20 days. Ironic thing is that if demand has come down some, even with gasoline inventory hit, forward cover situation may have improved of late, especially when seasonally adjusted. a lot of that is due to gasoline imports which likely can not be sustained. so between that and new refinery shut-ins???

 but the market is stabilising becasue both the similar graph for oil and distillates are above the 5 year average range.  if either had been low going into the hurricane season, you would see some other effects in those markets.

someone said that oil bought but still overseas awaiting import is counted in these inventory numbers. I do not think that is correct.  the weekly inventory reports posted are the same graphs just in more detail.

but the market is stabilising because both the similar graph for oil and distillates are above the 5 year average range



The crude stockpile situation in terms of forward cover is very healthy compared to 2004 and 2003 despite skyrocketing prices! so far stocks do not reflect a tight supply situation.

Well, I'm now completely confused as to what slice thru the supply chain and the books of the various entities involved this chart is showing :-).   Thanks :-).
This graph is showing for each week the Equivalent number of days of consumption for the crude oil stocks (in short 'Forward Cover'):

Forward Cover= (U.S. Weekly Crude Oil Ending Stocks Excluding SPR) / (U.S. Weekly Crude Oil Inputs into Refineries)

Slightly OT - has anyone else been following what's happening in Jo'burg @ the World Petro Conference. It appears to be an all out attack on Peak Oil (Exxon saying 3 Trillion out there to take, although did say "may have") and Matt Simmons (this from Saudi minister al-Naimi):

"We will soon be able to boost our proved oil reserves by 200 billion barrels using the latest technology,'' al-Naimi said today. ``We are further encouraged that there are vast areas of the kingdom that have yet to be explored. This leads us to say with confidence that Saudi Arabia's proved reserves will expand significantly in the years and decades ahead.''


"There will be plenty of oil available to meet future demand,'' al-Naimi said today in Johannesburg. Prices are high now because ``the petroleum industry faces infrastructure constraints and bottlenecks that are causing market volatility and restricting its ability to bring oil from the ground to the consumer.''

I presume these sentiments will be run ad nauseum by various Pollyannas, while little old ladies freeze this winter.

Hehe ... I saw that quoted in a CNN Money article and just posted the first part of it in another thread a few minutes ago. Looks like the "Pollyannas" are already quoting this. :-(

What I forgot to say in the other post I made is that after Simmons has already pointed out the faults in their existing reserve estimates, I'm surprised the Saudis would do the same old song and dance one more time by upping their reserve numbers. "Fool me once, shame on you. Fool me twice, shame on me."

I used to hear that saying and think of Scotty, from Star Trek.  Now I think of W stumbling through it.
If you think US stocks are low, here in the UK we have just
11 days worth of natural gas stocks with plummeting
production from the North Sea and inadequate facilities
for importing it.
Spent a lot of time on the EIA site re gasoline today. The stock they report is refiner stock only. What is referred to above as secondary and tertiary stock is not reported. If refiners can still operate but can't make shipments, EIA stock will go up regardless of what is happening at the pump. Also the drop in demand post Katrina that has been cited as evidence of price curtailing demand seems to have been almost entirely in the Gulf-Coast region. A 20% drop in that region, due to the hurricane, would account for nearly the total USA decline. Andy Weissman has pointed out that any stock saved in this way has probably been eaten up by the evacuation and return for Rita, especially Houston and Galveston. Projecting the present shut in and recovery forward for the next few months suggests to me that we will eat up maybe another 60 M of the 190 Mb still in inventory, by mid-winter. That will put us far below any absolute inventory level in decades, and at unheard of weeks of inventory. What price will be needed to destroy enough demand (5%?) to prevent that kind of shortage?  Note: if I'm right about the demand decline so far, recent price spikes have not dropped demand yet.  Murray