Fuel for America's transport needs

When groups such as OPEC try to predict the future demand for oil and petroleum products, one of the significant factors in that analysis lies in the demands of the US transportation sector. I have intermittently tracked those numbers, through the Vehicle Miles Travelled plot which the FHWA publishes each month, and here is the latest plot, which includes figures through last November. Bear in mind that this is a rolling 12-month total of miles travelled.

12 month rolling total of vehicle miles travelled in the USA (FHWA )

It is becoming increasingly evident that the plot has returned to a steady increase that has now been sustained at about the same rate as being maintained up to the earlier peak.

The question now comes, as evidenced by the remarks of London’s Mayor, as to how much higher the price of gasoline/diesel fuel will rise before it, once again, has an impact on overall growth. Overall growth was around 1.1% in November, up more than the annual average increase of 0.7%.

Driving is, to an extent, seasonally controlled, as this plot of the last three years driving, by month illustrates.

Travel on US Urban Highways by Month (FHWA)

Thus the bitter cold and poor driving conditions of the last couple of months may have had a negative impact on the overall totals, moving forward from the end of the plot, yet overall the recovery seems to be being well maintained.

This return to a higher level of demand is the subject of the front page comment on This Week in Petroleum which the EIA issued today (Wednesday). They note that transportation accounts for 72% of US petroleum consumption, at 12.9 mbd out of the current national total demand of 18 mbd. In the latest Energy Outlook the EIA has projected that GDP will grow steadily at 2.7% over the next 25 years, including a steady growth in transportation fuel use.

DOE projections for energy distribution (Energy Outlook )

As an aside it is interesting to note the EIA comment:

Although the situation is uncertain, EIA’s present view of the projected rates of technology development and market penetration of cellulosic biofuel technologies suggests that available quantities of cellulosic biofuels will be insufficient to meet the RFS targets for cellulosic biofuels before 2022, triggering both waivers and a modification of applicable volumes, as provided in Section 211(o) of the Clean Air Act as amended in EISA2007. The modification of volumes reduces the overall target in 2022 from 36.0 billion gallons to 25.7 billion gallons in the AEO2011 Reference case, equal to the AEO2010 Reference case.5

Well I don't think that is much of a surprise. In regard to more conventional ethanol production, it appears to have reached a current plateau at 0.9 mbd.

Turning to the more usual look at the charts, the anticipated decline in demand over the quarter has led to a drop in refinery activity.

Refinery inputs (EIA)

However the production of crude dropped even faster:

US crude oil production (EIA)

The difference has come from increasing imports. We will see whether this is a transient or more permanent change.



Retail gas prices now average $3.11 per gallon in the US (locally we are still hovering around $3.00), still a fair way below the $7.75 of the UK, but one wonders whether the steady ramp up in price will continue for another year or so, and if it does, what the consequences will be.

A VMT oddity and the latest TWIP

I have been writing about the use of the monthly reports from the Federal Highway Administration (FHA) on vehicle miles travels, at intermittent intervals over the past couple of years. It provides an informal way of trying to see how fuel demand is going in the United States, and perhaps some indications of the recovery of the economy. The figures are now out for September 2010 and I was looking at the plot I usually use of the rolling 12-month average of miles driven:

Rolling 12-month VMT (Source FHA )

I hadn’t been paying much attention to the lower scale, it shows years and it wasn’t until I tried to see when it was that the driving was last at this level that I realized that the scale does not include years that end in 4 or 9. Which is my oddity for the day.

Other than that the recovery of driving seems to be holding up both in urban and rural areas, and generally across the country. Texas seems to be doing a little better than most, but other than that the picture appears to be, as the plot shows, one of steady growth at a rate similar to that before the great oil price boost.

Moving over to This Week in Petroleum the picture similarly shows nothing particularly out of the ordinary. The EIA is going to take a slightly different look at storage capacity data, and that is their front page story of the week, but as one looks at the plots, other than the slow creep up in oil prices, there is little untoward in them. Domestic production has continued to increase, lowering the need for imports, at a time of year when refinery inputs in general are down.

Source EIA

At the same time ethanol production is continuing its steady climb in production:

Source EIA

At a time when the gilt seems to be wearing off the ethanol gingerbread, the public discussion seems to be having little effect on that reality.

But other than that, nothing much of significance that I can see, (which doesn’t mean that I’m not missing something – perhaps that gasoline demand has dropped to the same level as last year?)

September TWIP and July VMT

One of the metrics of the economy that serves as a marker to me on how we are doing relates to the use of vehicles, and as a consequence the amount of gas that is being used. Prices of gas haven’t changed much over the past months, nor has the price of crude. Thus without price fluctuations confounding the causes of change, it is possible to get some measure of how we’re doing from how much gas is being used. Recognizing that the summer driving season is now over, but that information is now available on how the patterns of driving went.

This Week in Petroleum, where I get the data on this, has, this week, focused on the impact that Canada is having on imports, given that it is the largest supplier to the USA at the moment. Part of the story they told this week was that the impact of pipeline problems between the two countries had not ultimately had much impact, and service is now resumed.

From TWIP Sept 29, 2010

The Canadian supply has remained relatively steady over the years, and not been much impacted by the declines in demand recently, though as a seasonal event, the demand for crude is currently falling.


If one looks at the TWIP monthly figures rather than the annual demand, the fall in demand that started some months ago, is continuing, and given that domestic production remains relatively level, it will be interesting to see when this turns around.


The amount of gasoline produced from this crude input would, logically, also fall, as it has been though it has just recently had a little uptick, relative to the trend of a year ago.


That is as a result of an increase in demand that can also be seen in the TWIP figures.


One week’s data should not, of course, be construed as having much import on its own, but it is worth watching.

Ethanol production, after a relatively steady, though small increase, concurrently had a slight dip, though this is, I suspect, likely to be insignificant in the longer term.


Looking at vehicle miles driven, the last report for which relates to July numbers, the curve (bearing in mind that it is a 12-month rolling accumulation) has picked up and is now past the early “bump in the road” which we saw earlier in the year.


The slope is not yet that exciting, with the overall levels still equivalent to those back in 2005, but it is upward and a recognition that things are doing better. And it appears to be an across the board increase around the country, and in both rural and urban driving.

The question however, will likely arise before too long as to what impact the increasing demand is going to have on prices. For while the demand in the US and Europe has seen anemic growth, that in Asia is much more robust, and has been consuming the “slack” that had been left in global demand. We shall see how this impacts the capabilities of world suppliers to continue to meet this, in the months ahead.

TWIP for August 11th

Before the Deepwater Horizon incident I would follow the EIA weekly announcements known as This Week in Petroleum and occasionally comment on what I thought to be important. At the same time, trying to discern how the economy was improving, if at all, I would look at the data from the FHWA on the vehicle miles travelled, and include that each month. For a variety of reasons it looks as though I haven’t done that since February at which point the US demand for gasoline was falling below that of a year ago, inputs to refineries were below that for 2009, and the VMT (which appears three months later) were showing that November numbers were somewhat more positive.

So with this 6-month hiatus, and being now towards the end of the summer driving season, how have things progressed?

In terms of the demand for gasoline, this is up close to 200,000 bd over last year:




Just to see how it progresses over the next few months, as corn and grain prices may rise due to the problems, inter alia, in Russia I am going to add the ethanol production curve from TWIP to the mix I will look at.


Gasoline comes from refineries, and looking at the input to those refineries, this has been increasing, relative to last year, though it may have peaked, a little later than last year, for this season.


Domestic production of crude remains quite flat, and for the next couple of months may depend on how relatively calm the Hurricane season remains, remembering that it was predicted to be more severe than usual.


Which means that the increase in demand must be met by increased foreign imports, which is what is being reported.


The increase in imports indicates that there is a growth in the economy that it is rising to meet, and this is recognized in the Short Term Energy Outlook that the EIA released yesterday. They anticipate that, overall, demand will rise by 140,000 bd this year, and by 170,000 bd next year.

With the global economy also growing, the question arises as to where this oil will come from. The EIA notes that while non-OPEC production is expected to rise by 0.72 mbd (million barrels a day) this year (based largely on increased production from the USA, Brazil and Azerbaijan) it will drop by 0.16 mdb next year (mainly due to falls in production in the North Sea and Mexico). Which, with an overall estimate of world demand growth being 1.6 mbd this year, and 1.5 mbd in 2011 raises the reasonable question as to where it is going to come from.

The only answer is OPEC (which, as I’ve mentioned before, is why it is their production which is now controlling the price of oil). The EIA see their spare capacity as being around 5 mbd. I personally think this is about twice the real value, but that is a discussion for another day. (But if the growth rate continues for 3 more years it may well be consumed).


So will growth rates be sustained? Well the main story at the front of the TWIP this week was on the rising demand for jet fuel, for the first time since 2007, and having just come from a total of 3 flights that were all “full to the gunwales” the demand for travel may be picking up.

But if one goes back and looks at the VMT for May, that recovery that I was beginning to see in February, hasn’t continued through the Spring, but rather reached a plateau. We will have to wait a couple of months to see whether that number has picked up as well as the air travel.

12-month rolling summary of vehicle miles travelled through May 2010 (FHWA).

Gas demand, miles travelled and salt for the roads

It has been a while since we looked at the TWIP and natural gas pages of the EIA, so (in order to maintain the historic record) a quick trip over there, and to the VMT page of the FHWA is in order. I will get to natural gas in another post, but the increase in natural gas prices in New York from $6 to $11 over the last week (it was as high as $14/kcf last Monday) is something eye catching. It is interesting to note that while domestic crude is about at the highest point it has been in quite a while (allowing for a little drop due to the weather and season), there is now also a slight uptick in imports as well, indicating that the domestic gains in production may no longer be enough to keep up.

Domestic crude production end of 2009 (EIA)

Crude oil imports end of 2009 (EIA)

Gasoline demand has been relatively stable , but if you go back to the seasonality of demand this is the time of year when gas prices have historically been lowest, as has demand. From now on the growth should be relatively slow but steady into June when it reaches the peak during the summer driving season. So how are we doing this year?

Gasoline demand at the end of 2009 (EIA)

The question becomes whether we will drop down in demand through February, as we did last year, or whether we will return to the more historic curve with the earlier minimum. We shall see. (Robert Rapier’s recent essay on the possible perils of making predictions is still in my mind).

So what is happening with the VMT? Well, bearing in mind that these reports run a little late (the latest is for last October, there was a y-o-y drop overall of 0.5% in miles driven, after having seen an uptick in driving for the past few months. Interstate driving has risen, but it is the traffic in the “other” urban that was hardest hit, with other rural also being down over 1% (arterial and interstates were less affected). And for the overall summary curve?

12 Month running total of travel on US Roads through October 2009 (FHWA)

Well it is flattening out a bit, and may well do so through the next couple of months, given the typical drop in driving in winter, which will be exacerbated this year by the colder weather and greater depths of snow that are curtailing driving at the moment.

One of the things that I pointed out in comments after yesterday’s post was a direction to the Youtube video of the Head of the British Met Office getting his head handed to him by the BBC, in an interview that connected three erroneous medium term forecasts (for last winter, last summer, and this winter) with a 25% performance based increase in his salary. There is a relevance to this that may not be quite immediately discernable – but basically when, in times that are financially tough, counties, cities etc look to their budgets for the year, they tend to take such predictions of future weather into account when then order grit and salt that can be used on the roads to improve driving in winters such as the current one.

Unfortunately for the second winter in a row the Met Office has that prediction wrong, and the problem that it has caused is that there is now not enough salt and grit to go around. Villages in Britain have been cut off for weeks without the county doing anything to help, and this is unlikely to change soon.

But County Councillor Keith Young, who has a responsibility for highways, gave residents little comfort, asking them to 'bear with us'.

He said: ‘At the moment we have a very extreme set of circumstances and the priority is to keep the main roads clear.

‘As soon as we can we will treat the other roads we will do but we will not jeopardise our grit stocks.

‘My message to the residents in Cow Ark is that we are not doing this deliberately and I am sure their community spirit will see them through.’

I am sure that is a great consolation.

The role that highways play in providing access and communication, whether of bread, fodder for sheep or tankers for milk is often underplayed in the needs of those that live in more rural parts, but it is equally critical to the overall national picture since the costs to those living in the city may become more evident as milk becomes less available in consequence.

Salt for highways is often produced from underground mines, it should be kept dry until used, and for best effect should be spread on snow at a density of around 20 – 40 gm/sq. m.. There were some lessons that could be learned from the harsh winter in the UK last year (pdf) unfortunately (as is now becoming evident) those lessons were not well learned and there is now a national shortage of salt to treat the highways, even as the cold is anticipated to continue for another couple of weeks or more. In the UK there is only one major supplier of salt, a mine at Winsford in Cheshire, and while they can increase production to a degree with demand, beyond a certain point that becomes impractical due to a lack of machines, operators and suitable transport to deliver it to the customer. Thus the criticality of counties knowing in advance how much they will need, and the results of the failure of the Met Office to perform as they should, given their claim to be the best in this business in the world. (Not perhaps something they should have been stressing against their current record).

In the USA salt comes from regional mines, but there is the same basic need for information, and the same reliance on forecasts to assess how much salt and grit to stock. It might not have been politically correct, just before the Copenhagen meeting, to admit that the Northern Hemisphere was going to have a severe winter against AGW predictions, but it should have been the correct step, had folk been concerned about doing what they are paid (and apparently very well) to do.

Seasonality of Demand

Well, with Thanksgiving, and a slight problem in transportation that got us home a day late, it is belatedly time to catch up a little on the latest TWIP report . As we look at reports of weak demand for gasoline and rising stocks, it is important to remember the context within which they are being reported.

Consumption of gasoline is somewhat controlled by season, as is overall oil demand.

Source EIA.

Demand therefore will normally decline in the winter months, and one can see this for the current gasoline demand plot:

Source (EIA )

Demand peaked in August and will now decline until late in February. (Although when, back in that time earlier this year, I looked at these curves I was unable to see a pickup in driving until after April). Looking at how the FHWA record of driving is progressing this month (bearing in mind that the running 12-month total is some months behind current). Overall driving across the country was up 2.5% in September on a year-on-year comparison, and this month there was a gain in all regions of the country. (All but the North-East showing a gain of more than 2%).

Vehicle miles driven reported for Sept 2009 (FHWA )

The changing demand for gasoline with the change in seasons, and the current drop is thus then reflected in the historic change in gasoline prices, which, when averaged from 1990 (taking the data from the EIA) gives:



This is just for regular gas (which is the first column in the table at the EIA that I have derived it from).

Prices have, on average, fallen to a minimum around the beginning of Christmas week, and peaked about the end of June (Morton Downey has a similar sort of chart in Oil 101 which shows that driving peaks at the beginning of August, on average, and is at a minimum in February.

If one looks at the last couple of years, from the EIA plot, one can see that there is, as with demand, a clear seasonality in price, which suggests that no-one should be unduly concerned over prices for the next two or three months, since they will likely fluctuate a little as a result of the normal fall in demand.

Gas prices over the last two years (EIA )

It will be interesting to see, however, what starts to happen as demand picks up, as it normally does, somewhere in towards the end of February and then more strongly in May. Because I suspect that it will be about then that supply might become a little tighter.

We have the Saudi’s at the moment agreeing to hold supplies to the United States at a constant volume, while they previously agreed to increase sales to China as both countries work to cement ties, and while the production from Manifa (h/t Leanan) is pushed back to 2015. Whether this will have any overall impact on the global market will likely become more evident as we move into the summer of next year.

TWIP this week focused on the change in ownership of the refineries in the United States over the past decade. It is best illustrated with this table that they provided.



As you can see, even though there are no new refineries, by improving capacity within existing plant, overall production numbers have increased. The footnote however recognizes the recent closing of the Delaware City refinery and the loss of 210,000 bd of refining capacity.

Gasoline demand, August driving and winter weather

Wednesdays are when the TWIP (This Week in Petroleum) is issued by the EIA. Today’s cover note deals with the relatively changing prices of distillates (which includes diesel) relative to gasoline. In the past year the United States has become a net exporter of distillate, and even though the market has fallen, continues to be one, averaging about 350 kbd.

Looking at the gasoline demand, as we move further into a winter heating season, with less driving, there has been a fall-off in demand, though it is still higher than a year ago.

U.S. gasoline demand (TWIP )

When one contrasts this with the changes in driving patterns, now available from the FHWA for August, overall travel is up some 0.7% over August of 2008. Looking at the curve above the greatest impact of the recession on gasoline demand did not occur until September last year, and so this shows a further increase in the recovery. That is best shown by the running 12-month of vehicle miles driven.

12 month running total of vehicle miles driven in the USA. (FHWA)

The upturn in the curve is now clearly defined. The gains are not, however, this month distributed around the USA. Both the North East and the North Central parts of the country are showing a decline in driving, and it is in the South and West that there have been the gains, with Texas and the South Gulf showing the largest increase of some 1.7% over last year. Both urban and rural driving are still showing gains over last year, though now falling away from parity with 2007 (which rural driving almost matched for a short time in July).

We are left to await the coming winter, with the debate as to how hot it will be, relative to normal, given the El Nino condition. Going to the NOAA Winter Outlook prediction site, the winter is largely predicted to be warmer than usual across the country, and as a general rule perhaps a little drier in some parts, though wetter along the coasts. That may help (?) encourage driving over burning winter fuel – we’ll just have to wait and see how it plays out.

Winter outlook for temperatures.

Winter outlook for rain and snow.

These predictions are relatively-short term and it will be interesting to come back to this page in the Spring, and see how accurate the predictions were. (Likely better than mine from 30-years ago, to which I hope to return, perhaps tomorrow).

Gasoline consumption and today's TWIP

So how far has the recovery progressed? Not being one of the many financial gurus that have been peering over the bull’s entrails for the past six months my measure, as you may have noted has been a bit simpler. I have been looking, each month, at the amount of miles driven and the gasoline consumed, as reported in the weekly TWIP, by the EIA. It is not a particularly accurate measure but it does indicate a level of activity that first started reversing the decline last April, and it has been picking up ever since. The current curve, published today, shows a continuing increase in consumption.

U.S. gasoline demand (EIA TWIP)

However it does require a little clarification since the data that the TWIP is plotting (if one goes one layer deeper into the site) appears to be the finished gasoline supplied. However, since this tends to fairly rapidly transfer to the consumer in most cases it remains a simple guide to what may be happening.

So the curve is going up, and so I wondered where we stood relative to the conditions of the economy pre-recession. Going to the tabulated data – conveniently in a spreadsheet – I simply sorted the data by the size of the demand (thousands of barrels a day).
The data for the top weeks of production shows that we peaked (in the short term) in 2007.

Top 20 days of gasoline production/consumption in the United States (EIA)

However, if we look at the second block of 20:

Second top 20 days of gasoline consumption/production in the United States (EIA)

You will notice that in the week of May 22 (number 27) and of August 28 (number 40) 2009 has made it back into the list. This is not the season for high demand, and so, even though the trend is up, the next 2009 entry does not come until number 85, but the fact that the numbers have headed back, and that the difference between the current peak (9,762,000 bd) and the highest this year (9,538,000 bd) is only 224,000 bd might suggest that the public awareness of the OPEC restriction on production – or the relaxation of that quota system – may come a little earlier than we might otherwise have thought. The FHA plot for the month on actual vehicle miles driven for August is not out yet, but the picture for May was showing that, barring the South East, the rest of the nation was starting to drive more than in the same period of 2008.

Change in regional driving habits over 2008 for May of this year (FHA)

It will be interesting to see how the winter turns out, though the use of different fuel sources, the economy and the type of winter we’re going to have are all still in question.

Interestingly the commentary at the start of today’s TWIP dealt with the need for more transparency in the global oil market, noting that information on inventories is the most opaque. Since this is the buffer between production and consumption and the growth of “floating storage” – an even more opaque value – means that the real trends in production can be masked and even more difficult to predict.

Driving is picking up, but is that good?

Having effectively been away for a couple of weeks, as good a place to start as any, to catch back up with the current state of things, is This Week in Petroleum, which is looking, this week, both at the changing locations of our suppliers of crude, as well as noting the current status of the industry.

The opening review looks both at the change in the relative sources of our crude from 2000 to 2008, and then the more immediate changes over the last quarter as OPEC has acted to tighten supply against reduced demand. I am going to insert two of the relevant graphics here to show the changes. Following that I want to talk about changes in driving habits, not only here but also in China and India, because I begin to suspect that we are going to be significantly under-estimating the increasing demand for crude on the world market that those countries are going to be placing. That will tighten demand against supply, faster than I believe we are currently expecting, with a consequently faster ramp up in prices than DOE is currently projecting.

The first picture shows the changes in the sources of crude over the last 8 years:
Source TWIP EIA 24 June 2009

And the second the changes in imports this year:
Source EAI TWIP 24 June 2009

Currently US inputs to refineries are, while steadily increasing, down around 500,000 bd over the same time last year. And if one smooths both curves by eyeball, its seems that gasoline demand in the US is running only a little down on last year.
Gas demand through 24 June 2009, (Source TWIP.

Now if we wander over to the latest travel info from the FHWA, which is the April Report you can see that there is now an uptick in the curve over last year, and in all regions of the country travel increased, with the national average going up 0.6% (1.4 billion vehicle miles) over the same period (April) last year.
Vehicle miles travelled (Source FHWA)
The curve does not show the full uptick yet, since it is a 12-month rolling average.

So American driving habits are beginning to pick back up. This will lead to some increase in demand, but we need to add a couple of additional caveats that are likely to impact world demand, and which I have posted on before, but will briefly add back here to explain why I think that the world demand:supply situation is going to tighten faster than most believe, and thereafter why prices are going to be back up a lot faster than the Administration (among others) anticipate.

The first is the demonstrated success of the Tata Nano in India. In a country where the demand for cars has been about 1.5 million a year, this is selling at 100,000 a month. Demand for gas won’t be immediately felt since the factories are only geared to selling about that many a year, but you can be sure in a year or so that the burgeoning demand will be met, and the gas demand, from families for which this will be the first car, will also rise accordingly.

Switch over to China. The Chinese Administration is not running dual carriage ways up through river valleys because they look pretty. This is to meet the growing demand for cars, and the pathways along which to drive them. Chinese car sales are booming.

GM says China now accounts for nearly 25% of its global sales. With no end in sight to the troubles in the U.S. auto market, GM, Ford, and Chrysler likely will become increasingly reliant on China's still largely untapped market, says Yale Zhang, a Shanghai-based analyst for CSM Worldwide, an auto industry group.

"If you want to grow your overall volume, this is where you need to invest," Zhang says.

China is on track to sell 11 million vehicles this year, according to the China Passenger Car Association. That would be up 17% from 2008, and a stunning 20 times the number of vehicles sold in China just a decade ago. Zhang says this year China likely will overtake the USA, where expected sales are around 10 million units, and become the world's biggest car market for the first time.

China's 1.3 billion people "are simply wild about cars," says Michael Dunne, a Shanghai-based managing director of J.D. Power and Associates, an auto industry group. He says the surprising strength of China's auto market has been driven not just by economics, but also by a kind of psychological shift that has come with prosperity.

"There is the thrill of individual mobility, going from point A to point B in their own time, and on their own terms. But it's also an opportunity to declare and project their own success," Dunne says.

They really don’t start taking vacation in China until the end of this month and already the refineries are gearing up for the anticipated demand. And with China seeking to ensure its supply by acquiring others the world market will tighten, and the amount available for the rest of us is going to decline.

It is hard to see how this cannot accelerate the coming crisis in supply, and even more forcefully hand OPEC the keys to the world’s wealth.

A quick look at gas and miles travelled

Having missed both This Week in Petroleum, and the monthly release of miles driven, I thought I would just pop these up. Partially because we are seeing a slight pickup in miles driven, in parts of the country, which slowed the drop. (But bear in mind that this plot includes information only up through February.

February miles travelled

Let me just look in more detail at the end of the graph.

Top part of the Curve showing the change in slope.

If we look up at the top, the declining slope of the line is beginning to turn up, a little. When we contrast this with the data from TWIP on gasoline demand, one can see that this is remaining sensibly flat.

(Source EIA)
While the historic pickup that usually applies from April through June is not evident, the slope does appear, if anything, a little positive. But it remains early to make predictions, or do anything much beyond remaining optimistic.