In response to Petroleum And Gasoline Usage Charts for June, July, August; Unemployment vs. Gasoline Usage Analysis, a post based on weekly petroleum stats from reader Tim Wallace, I received a very nice email including a superb set of charts from James Beck, Lead Analyst, Weekly Petroleum Supply Team for the Energy Information Administration.
James gave me permission to use his name and his charts as long as I mentioned that his email reflects his personal opinions, not necessarily that of the EIA.
It is a pleasure to get an email from a government worker who takes his job seriously, is exceptionally knowledgeable on his subject, and is willing to be quoted by name.
James writes ...
Hello, Mike and Tim,
Thank you again for using the data from the Weekly Petroleum Status Report (WPSR) in your analyses of the demand of petroleum and gasoline. As the Lead Analyst for the WPSR at the Energy Information Administration, I always appreciate when others use our data in providing analysis!
I have updated my charts that I sent to you a few months ago. I have included Total Petroleum "Product Supplied" (a proxy for demand), Gasoline Product Supplied, Total Distillate (Diesel and Heating Oil) Product Supplied, and Kerosene Jet Fuel Product Supplied charts for you to review. These charts (with all of their data included) are based on the EIA's Petroleum Supply Monthly (PSM) data. The reason to look at the monthly numbers is that they are more reliable than the weekly as the survey is of the entire industry and there is a great deal of extra time used to verify the data. Many people believe that the monthly numbers are a revision of the weekly numbers. This is not true. These are separate surveys. Where the monthly surveys the entire industry and collects much more detailed information, the weekly information is based on a sample of the industry drawn from the monthly reporters, collects less information, and is focused on timeliness versus completeness. The weekly numbers are estimates of the most recent week's data based on the sample and are a snapshot in time. The weekly is a very good indicator of the data, but the monthly is the touchstone (at least until the Petroleum Supply Annual (PSA) is released--which is, in fact, a revision of the monthly data). Also, the monthly numbers from 2011 have been revised (as have some of the numbers in 2012) with the release of the PSA late last month. You can see the revisions in the "Data 1" tab of the attached spreadsheets.
In each of these workbooks, I have also shown a comparison of the weekly data versus the monthly data. This shows that the weekly does do a reasonably good job of capturing trends; however, sometimes these are exaggerated in the weekly data. Although the monthly is lagged by two months, it is the better measure of the full picture.
The following commentary is from me as a private citizen, not as a spokesperson of the Energy Information Administration or the Department of Energy:
The data support your general point that total petroleum product demand is at 1997/98 levels. The running three-month average that I am using (Apr/May/Jun--the last three months available) show that total demand has bounced above the lowest point for the same three-month period in 2009, but remains significantly below 2010 and 2011 levels--remaining very near 1997/98 levels. This 15 years of demand destruction cannot be explained fully by increased efficiency or increased use of biofuels and renewables (these have, at most, a marginal effect). This is truly an indication of the real and continuing trouble in our economy, high unemployment and underemployment, loss of manufacturing, and reduction of shipping. Total product supplied for April - June has averaged 18.652 million barrels per day, 0.5% above the same period in 2009, and is the second-lowest level for the three-month period since 1997 (which was at 18.487 million barrels per day).
Demand for gasoline continues to be below 2002 levels and the lowest level for April - June since 2001 (earlier this year the PSM numbers had shown a slight increase over last year; however, with the revisions to the 2011 numbers in the PSA slightly upward, this trend was shown to be incorrect). Gasoline demand had rebounded somewhat in 2010 (rising near 2004 levels after the recession in 2008), but has fallen below the recessionary levels of 2008/09 in the last two years. At 9.035 million barrels per day, gasoline demand is down 0.4% from last year, but down 4.8% from the April - June peak in 2007 of 9.491 million barrels per day.
Distillate demand, on the other hand, continues to show weakness. For the period of April - June, it is down nearly 4.5% from last year. At 3.729 million barrels per day, it is at its lowest level since 2002. The concern I have is the year-over-year decline this year. Since diesel demand is a very good proxy for the health of the economy (all shipping uses diesel--trucking, rail, barge, etc.), this weakening from last year continues to be source of concern for the economy.
Demand for jet fuel has also fallen dramatically from 2007/08 (it had also fallen dramatically after the 9/11 attacks, never fully recovering to the levels seen from 1999 - 2001). At 1.437 million barrels per day during the period from April - June, KJet demand continues to be at levels we have not seen since 1994/95. The 2012 level is the second lowest for this three-month period since 1995. Although KJet demand is up 1.2% from the 2009 low for the period, it is down 2.4% from last year.
These numbers do not tell me that we are in a recovery. Despite increases in distillate and KJet demand in 2010 and 2011, and in gasoline in 2009 and 2010, these were well short of recovering from the decline in 2008/09. The decline year-over-year in these three core transportation indicators suggest a slowing in the economy if not a recession.
I hope you can make use of the charts. Please let me know if I can be of further assistance.
Lead Analyst, Weekly Petroleum Supply Team
Energy Information Administration
Office of Petroleum and Biofuels Statistics