UPDATED: see below
A few days ago I did a report on the U.S. Energy Information Administration (EIA) numbers for carbon dioxide emissions, showing that it was clearly down, and back to levels not seen since 1994, and noting that since Kyoto in 1997, U.S. emissions have dropped even though Kyoto was never ratified in the USA.
If you thought that was troubling and strange have a look at these numbers which also indicate the state of the U.S. Economy. First, the number of miles driven monthly for the last 30 years. As you can see, just like global temperature has flatlined, so has the number of miles driven.
Source data: http://research.stlouisfed.org/fred2/series/M12MTVUSM227NFWA
Now the amount of gasoline sold. Note the regular seasonal “heartbeat” pattern up to about 2008, then that pattern gives way to a precipitous drop at the end.
Source data: http://www.eia.gov/dnav/pet/hist_xls/A103600001m.xls
If that doesn’t paint a grim picture of the U.S. economy, I don’t know what will.
Zerohedge writes:
…but the biggest question we have is just how did the biggest boost in energy and engine efficiency occurred at two key junctions: Just after the Lehman Failure, and just after the US downgrade and the first debt ceiling crisis, when the total sales of gasoline by US retailers literally went off the charts, and which data series is now languishing at levels not seen since the 1970s (unfortunately we can only estimate: not even the EIA’s data set goes back that far).
Perhaps, just perhaps, Occam’s razor applies in this situation as well, and the collapse in energy demand in the US has little to do with MPG efficiency, higher productivity, and throughput mysteriously achieved just when the entire economy was imploding in the months after the Lehman failure, and despite the re-emerging proliferation of cheap Fed debt funded SUVs and small trucks, and everything to do with the US consumer being slowly but surely tapped out?
Of course, if that is the case, than the US economy is far, far weaker than even we could have surmised, although it certainly would explain the desperation with which the Fed is doing everything in its power to preserve the levitation of the S&P, i.e., the confidence that all is well despite all signs to the contrary. Because should the market finally be allowed to reflect the underlying economy – not the administration represented economy, but the real one – then everything that has transpired in the past five years will be child’s play compared to what’s coming.
I wonder if that brilliant economist of the NYT, Paul Krugman, can pull the wool out of his eyes long enough to comprehend this?
h/t to Kate at Small Dead Animals for getting me interested in this enough to plot the data myself to see if it was true.
UPDATE: I added this is response to comments about the number of miles not dropping as fast. “jeez” points out that miles driven are an estimate from surveys.
If people are driving less miles, we have less consumption, and that would mean excess supply and lower prices. Lower prices should then result in more people driving more, sort of a self correcting feedback.
Instead what we have is a 50% drop in retail sales of gasoline during a period of reduced driving.
That says to me that many people have just stopped buying gas. Consider that 90 million people are now out of the workforce. Look at this graph and that helps explain part of what we are seeing.
UPDATE: Correction. From this comment, I agree, the Zerohedge article focus on retail sales is misleading, see new plot I did below. I’m not privy to the vagaries of gasoline supply/sales channels, and had I been, this would have raised more suspicions. Thanks to WUWT readers for the peer review! – Anthony
Anthony,
As a few others have mentioned, the bug is in “retail sales by refiners.” There has of course been wholesale in the past to off-brand distributers (i.e. 7-11 selling gasoline that they sure don’t refine) compared to Exxon selling Exxon refined gasoline. At those drop-off points what likely happened is that fewer people were willing to spend a few extra pennies stopping at Exxon, and now buy their gas at Wal-Mart or Kroger when they do their grocery shopping.
The fact that
http://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=PET&s=MGFUPUS1&f=M
http://www.eia.gov/totalenergy/data/monthly/pdf/sec7_5.pdf
both align with the CO2 and other data (like total petroleum consumption) makes it much more reasonable to think there has been a ~10% decline in gasoline purchases than a 50% decline. Otherwise we would have to ask how we cut 25% of our Carbon use (petroleum is ~1/2 of our carbon use, and a 50% decline in that would be a total of 25% of all carbon) while only decreasing carbon emissions by ~10%.
(Note: To test this I plotted the EIA data below from here: http://www.eia.gov/dnav/pet/hist_xls/MGFUPUS1m.xls – Anthony)
A 10% decline would then be appropriately explained by 4% decline in labor, increases in fuel efficiency, and smaller factors like online shopping (remember, somebody still drives it to your house – and usually they leave a large truck idiling while they walk the package up and have you sign). A 10% drop is still a huge amount of gas, but it is not the same as a total societal collapse that a 50% drop in 4 years would indicate.
I assume this was probably an honest mistake, but since it has been pointed out several times I think the most honest thing to do is change the data set and correct the article.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
A very real hockey stick
Is there a Private Sector is Doing Just Fine graph?
Something seems funny here, at least to me. I would have assumed that the number of miles driven and the number of gallons of gasoline sold would be proportional, save for an adjustment that reflects the long-term improvement in average fuel economy over time. But these plots seem do not bear that out. Since 2008, the number of miles driven has interrupted its upward trend and flatlined, while the amount of gasoline sold has decreased by a factor of 2. What am I missing?
Not a hockey stick. It is a scythe:
http://www.bishop-hill.net/blog/2013/4/1/introducing-the-scythe-josh-230.html
How are we driving in miles at approx 2004 levels but buying half the gas? Something doesn’t look right.
The number of miles driven might have dropped because more business is done online.
My business partners and clients, I hardly speak to them anymore – when I’m not typing into skype or phoning them, emailing them, directing them to look at web presentations I’ve prepared, I’m usually asking them to download something.
With all the new productivity tools which have arisen over the last decade, you barely have to get out of bed in the morning to have your business at your fingertips – in fact, on days when I sleep with my phone under my pillow, so I can take support calls, I don’t.
Better check your hole cards on that retail sales chart. I question the drop in 2012.
I think you need to look at this http://www.eia.gov/dnav/pet/xls/PET_CONS_PSUP_DC_NUS_MBBL_M.xls which includes all finished gasoline sold. I am not sure, but I think your chart leaves out the reformulated gasoline.
1) I really dislike graphs that aren’t based at 0.
2) @MDR the collapse of trucking and SUVs? That would bring down gallons faster than miles driven. Though looking at the data it’s “retail” gas, not wholesale, so I suspect the sample isn’t quite the same. The zerohedge post goes more into detail.
MDR says:
April 9, 2013 at 9:35 pm
But these plots seem do not bear that out. Since 2008, the number of miles driven has interrupted its upward trend and flatlined, while the amount of gasoline sold has decreased by a factor of 2. What am I missing?
This is just an educated guess on my part. But could it be that places like Los Angeles has fewer cars on the road so less time is spent in traffic jams and more of the gasoline is actually used to drive instead of idling?
CO2 tax is not the Tax per say, it’s about control. Picking winners and losers and rent seekers.
MDR
One quantity is easy to measure. Fuel producers KNOW how much gasoline they sell.
Miles driven can only be vaguely estimated from surveys (until every vehicle is tracked-coming soon.)
Looking at the other comments, I do think something is wrong here also.
@jeez if people are driving less miles, we have less consumption, and that would mean excess supply and lower prices. Lower prices should then result in more people driving more, sort of a self correcting feedback.
Instead what we have is a 50% drop in retail sales of gasoline during a period of reduced driving.
That says to me that many people have just stopped buying gas. Consider that 90 million people are now out of the workforce. Look at this graph and that helps explain part of what we are seeing.
I have to agree with jeez above. Miles traveled/retail sales cannot be reconciled.
This is staggering. Halving retail sales goes way beyond economic issues.
It says societal collapse.
Just talked to a statesider. He asks:
1. Do the numbers measure pure gasoline or do they include gasoline cut with ethanol?
2. Do the trends take into consideration the increased use of natural gas?
3. To what extent are the fuel efficiencies mandated by state laws such as California taken into consideration?
It looks like reformulated is included in these numbers
http://www.eia.gov/dnav/pet/pet_cons_refmg_d_nus_VTR_mgalpd_m.htm
Each of the tabs on this link show different cuts of the data, refiner owned, wholesale sales, etc. In Jan 2013 looks like they sold 27 Million gals….the lowest monthly since…well, this chart only goes back to 1983
http://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=pet&s=a103600001&f=m
boy, i bet they are *really* happy they realigned into the chips and soda business.
A lot more people running on empty
Anthony, it maybe doesn’t change the shape of the curve much but, unless my reading glasses are letting me down, or there’s something I don’t understand about “sales by refiners”, the y-axis numbers are off by a factor of 10. Total annual US consumption topped out at 140 Billion gallons in 2007, so a daily average would be 1/365 of that, or around 10X of the y-axis numbers.
http://www.eia.gov/tools/faqs/faq.cfm?id=23&t=10
http://www.eia.gov/dnav/pet/TblDefs/pet_cons_refmg_tbldef2.asp
and, also, here”s the definitions they use so you can see exactly what is included.
@ur momisugly JC, correct. The US economy is collapsing in slow motion. Why else would we see 40% of the normally working population NOT WORKING. Get a clue – record levels of US sovereign debt (17 trillion), record levels of people on food stamps, record levels of people taking early social security, record levels of people on disability, record levels of people at poverty levels, record levels of homes foreclosed, record low levels of home sales and new home construction…the list goes on and we are just waiting for the other shoe to drop.
The second the Fed stops pumping money or the interest rates start rising the fat lady will be taking a deep breath. If something can not continue to go on — it won’t.
PS to my above post:
Maybe refineries are a fraction when also adding in:
“blending plants, pipelines, and bulk terminals.”
But overall US retail sales have increased by 9.58% over the last two years: http://ycharts.com/indicators/retail_sales/chart#series=type%3Aindicator%2Cid%3Aretail_sales%2Ccalc%3A&format=indexed&recessions=false&zoom=5&startDate=&endDate= . Could it be that people no longer drive to Walmart because they order what they want on line and wait for it to be delivered?
J Barber says:
April 9, 2013 at 10:58 pm
But overall US retail sales have increased by 9.58% over the last two years: http://ycharts.com/indicators/retail_sales/chart#series=type%3Aindicator%2Cid%3Aretail_sales%2Ccalc%3A&format=indexed&recessions=false&zoom=5&startDate=&endDate= . Could it be that people no longer drive to Walmart because they order what they want on line and wait for it to be delivered?
*************************************************************************************************
Or the figures are fudged???
I don’t believe there are less miles driven in my area. I’m always astounded at the constant high level of traffic all day, every day. The roads are packed at all times, especially midafternoon during the week when a fraction of people are supposedly at work.
I live on a small side street that serves a community of a couple dozen homes. The road would not really be used as through traffic. And yet every day, all day (even just now and it’s 2:15am ! ) a car goes past on average every 3-5 minutes. Where is everybody going all the time? The same cars go up and down past the house all day , every day. It’s worst when the weather is nice and sunny. I’ve definitely noticed that temperature increase is correlated with a dramatic increase in traffic.
There is still something funny about the original two plots.
I still don’t know exactly what I am looking at, and consequently I think the analysis about the degree to which the labor market influences these numbers is premature.
As Chris [April 9, 2013 at 9:41 pm] stated succinctly: How are we driving in miles at approx 2004 levels but buying half the gas?
As philincalifornia [April 9, 2013 at 10:43 pm] stated: The y-axis numbers [in the “total sales by refineries” plot] are off by a factor of 10.
My take on Chris: Fuel efficiency hasn’t improved that much in 9 years, and the amount of transportation fueled by CNG or electric remains miniscule in comparison to gasoline powered vehicles and thus cannot not account for much of the 50% drop. It’s possible the survey of miles driven is at fault.
My take on phil: But that leaves the other plot, and total sales by refineries would seem to be easier to measure [and thus more accurately measured] than miles driven. However: Given that phil’s link states off the top that in 2011 the US consumed “a daily average of about 367.08 million gallons” of gasoline [side note: are 5 significant figures really necessary?], this does not match the daily numbers plotted in the post, as phil mentioned. Yes, I know the table says monthly, but that refers to the sampling [of once per month] and not to the consumption rate divisor [which is per day]. This is verified by the first link given by rk [April 9, 2013 at 10:38 pm], where the exact same data are provided in the “1983-2013” link in the “motor gasoline” row of the table, down to all of the wiggles, in which the units are clearly “thousand gallons per day”, i.e. same units as the plot in the original post. That, in turn, causes me to wonder whether the total refinery sales plot is measuring some subset of the total amount of gasoline used for transportation, since it’s so much lower than the aforementioned 367.08 million gallon figure, and thus we are not being provided with the complete picture. I can’t find any indication of what this is actually plotted on the various source webpages right now, though, so I’m at a loss, especially since “gasoline” is fairly specific and primarily used for transportation in the US [though it is also used in agriculture as well].
Bottom line: Something still seems funny here.