Mark Anthony's  Instablog

Mark Anthony
Send Message
Mark Anthony, is an IT professional and who had a scientific research background before joining the information revolution. Visit his blog: Stockology (http://stockology.blogspot.com/)
My blog:
  • EIA Weekly Natural Gas Updates Indicate No Coal To Gas Switch 36 comments
    May 14, 2012 3:59 AM | about stocks: Q, PCXCQ, ACIZQ, BTUUQ, ECA, WLT, ARLP, CHK, CLF, COP, APC, EOG, DVN, BHI, KOLD, UNG

    Just a quick update. I have carefully analyzed archives of the EIA weekly natural gas updates, extracted useful information, and deduced accurate weekly NG demand figured to allow me to plot the data out:

    (click to enlarge)

    Here is the chart of weekly natural gas demand. Note that when the monthly tallies are deduced from the weekly numbers, they add up precisely to the monthly data that EIA releases:

    (click to enlarge)

    Seeing the above chart, where do you find natural gas demand in the electricity power sector? You probably need a magnifier to find it.

    [Update: May 13,2012] The updated chart is shown below:

    (click to enlarge)

    There is NO coal to natural gas switch. It's totally debunked. It does not show in the data. It does not show up in the chart.

    BTW it is worth being a bit skeptical to some of the EIA data. Note this example below:

    (click to enlarge)

    The data was provided by BENTEK. I can't help shaking my head!

    I have submitted a formal SA article in which I discuss my findings in the EIA weekly natural gas updates. Please stay tuned!

    Update: Please look at the April monthly tally of NG usage in the power sector. The data in the data table only tallied 25 days of April, up to April 25, 2012, for a total of 436.45 BCF. The week ending in April 25 saw 119.613 BCF consumption for 7 days. Extrapolate that for 5 days will be 85.438 BCF, bring the whole month of April to 521.888 BCF.

    That would be an April, 2012 LOWER than the April of 2011 (at 525.956 BCF), by 0.77%.

    http://www.eia.gov/dnav/ng/hist/n3045us2m.htm

    So the alleged coal to gas switch so may people talked about, is already ENTIRELY GONE in April.

    Disclosure: I am long JRCC, PCX, ACI, ANR, BTU.

    Stocks: Q, PCXCQ, ACIZQ, BTUUQ, ECA, WLT, ARLP, CHK, CLF, COP, APC, EOG, DVN, BHI, KOLD, UNG
Back To Mark Anthony's Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

Comments (36)
Track new comments
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    You need to compare it to the year before, not to itself in the same year because there's a seasonal component to it. Imagine you were doing the same type of analysis regarding retail sales, you'd be saying after Xmas "look, retail sales are plunging!" even if they were growing 10% versus the same period the year before.
    27 Apr 2012, 08:38 AM Reply Like
  • Mark Anthony
    , contributor
    Comments (3595) | Send Message
     
    Author’s reply » Paulo:

     

    Compare with a year ago is illegitimate precisely because of seasonality: That's two very different winterand can not be compared. Each of the Y-o-Y percentage is off a completely different comparison basis, rendering the data totally useless to tell you anything. Week by week variation at least tells you precisely whether things are going up or down. And as I show, correlated with the monthly data they tell you precisely what the demand numbers are.

     

    The numbers shown on my data table are scientifically accurate. They agree with the monthly data from EIA, they are consistent within themselves.

     

    Most importantly I did NOT torture any data. There is only ONE WAY to set the parameters correct to agree with the monthly data. If I change any number just a bit they would not tally up to agree with the monthly numbers. I have ZERO flexibility to tweak the number to make them the way I want them to be, except for making they what they really should be.

     

    It might be a torture to you, Paulo, due to your lack of math skills.
    27 Apr 2012, 10:08 AM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    The data is showing huge increases even off the winter, and besides, the winter was UNFAVORABLY warm, so at the very least it wouldn't increase the growth rate, it would DECREASE it, which means that the switch that's happening is even DEEPER than the huge growth rates show.

     

    You clearly don't lack the math skills, Mark - what you lack is seemingly, intelligence. eheh
    27 Apr 2012, 10:38 AM Reply Like
  • Robin Hewitt
    , contributor
    Comments (5554) | Send Message
     
    "the winter, and besides, the winter was UNFAVORABLY warm, so at the very least it wouldn't increase the growth rate [of dispatch switching], it would DECREASE it"

     

    That's incorrect, Paulo. The warmer weather meant that residential use of gas was dramatically lower, more than offsetting the increase in the amount used by utilities. The residential demand drop lowered price for natgas, which in turn led to increased dispatch switching. Your denial of that weather link is leading you to make incorrect projections.
    27 Apr 2012, 11:58 AM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    Duh, I KNOW that residential, commercial and even industrial demand fell because of the weather.

     

    But we're talking about what happened to UTILITY demand, given that all the others will normalize over time. The coal to gas switching pertains ONLY to utility demand.
    27 Apr 2012, 12:04 PM Reply Like
  • Robin Hewitt
    , contributor
    Comments (5554) | Send Message
     
    You're misrepresenting my post, Paulo. The dispatch switching is price driven. If natgas prices fall, dispatch switching increases. Prices of a commodity fall when the demand for that commodity fall. Demand for natgas fell relative to its seasonal norm thanks to the warm weather. Utility demand is linked to overall demand because dispatch switching is driven by commodity prices and commodity prices depend on overall demand.

     

    A mild winter lowers overall natgas demand more rapidly than it lowers overall coal demand because the residential use of natgas for heating is a substantial component of natgas demand whereas use of coal for residential heating is essentially non-existant.
    27 Apr 2012, 12:14 PM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    Yes, and?

     

    I don't NEED to know why natural gas prices fell (though I do know it was both weather related and production related).

     

    What I NEED to know is that once relative prices get the way they are, there IS dispatch switching, leading to lower coal consumption and higher ng consumption, which might well bring natural gas prices up again and terminate the dispatch switching.

     

    I never said that weather was not part of the reason natural gas prices fell - indeed, I WROTE an article saying that weather was pushing them down.
    27 Apr 2012, 12:25 PM Reply Like
  • Robin Hewitt
    , contributor
    Comments (5554) | Send Message
     
    Paulo, the reason you need (or if you prefer, NEED) to consider the impact that residential has is because you wish to compare yoy and project that as a trend into the future. However, that sort of trend projection is only valid if you assume there are no special considerations for one of two years you're using. Since there IS a special consideration for one of those years, the trend projection is invalid.
    27 Apr 2012, 12:36 PM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    I project things like residential demand not based on the depressed 2011 weather, but on 2010's more normal weather with a 0% delta.
    27 Apr 2012, 12:40 PM Reply Like
  • Robin Hewitt
    , contributor
    Comments (5554) | Send Message
     
    You can't separate residential demand from dispatch switching. The latter is affected by the former and you're neglecting that.
    27 Apr 2012, 12:42 PM Reply Like
  • jsingular
    , contributor
    Comments (28) | Send Message
     
    Mark-I follow many of your articles and posts. Great work. I also like to dive in and digest the data at EIA, which gives a much different picture than many of the discussions taking place here on SA. Keep up the research and vigorous discussions, from this rookie investor who is focusing on energy commodities - I like your style.
    27 Apr 2012, 08:43 AM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    Keep in mind that what he is doing there is intrinsically wrong. He's doing it to keep believing in his misguided theory that there's no coal->ng switch happening because of the prices. Something which is entirely at odds with reality.
    27 Apr 2012, 08:50 AM Reply Like
  • Titan06
    , contributor
    Comments (301) | Send Message
     
    well industry says its happening
    but
    data says its not

     

    strange issue... however even if it happening it will not last long. Any uptick in gas and all bets are off as to coal to gas switching
    27 Apr 2012, 09:03 AM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    The data says it's happening as well - massively. Mark just tortures the data to make it say what he wants to hear. Read my first answer, it explains the error here.
    27 Apr 2012, 09:11 AM Reply Like
  • Robin Hewitt
    , contributor
    Comments (5554) | Send Message
     
    Read my reply to Paulo's comment above. He either doesn't understand or is in denial of the link between weather and dispatch switching.
    27 Apr 2012, 12:01 PM Reply Like
  • Titan06
    , contributor
    Comments (301) | Send Message
     
    lets see if mark covers it in his new article coming out soon, im interested to see year or year comparisons
    27 Apr 2012, 09:40 AM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    He's already done that, by comparing the weeks to the xmas week ... eheh. Falling into the SAME trap over and over again, just so he doesn't have to recognize that natural gas on a yoy basis is getting burned at much higher rates.
    27 Apr 2012, 10:03 AM Reply Like
  • Mark Anthony
    , contributor
    Comments (3595) | Send Message
     
    Author’s reply » There was no report on Christmas week. So I needed to fill in one weekly percentage change. But this is totally legitimate:

     

    1. I have two numbers that I must get correct: The Dec. 2011 monthly tally should match the EIA's monthly number; The January monthly tally also needs to match the EIA monthly number correctly.

     

    2. I have two parameters I can adjust to match, the starting week level, and that one missing percentage number in Christmas day.

     

    All scientist students will tell you this is totally legitimate. Two adjustable parameters to fit two confining conditions. You will only have on correct answer. There is no arbitrarity here. This is not data torturing. This is the correct way to calculate the data.

     

    Anyway the data conclusively proves that the so called switch is totally GONE in April, as NG usage in April, 2012 is actually slightly LOWER than April, 2011, by -0.77%.
    27 Apr 2012, 01:56 PM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    Yes, it's legitimate - we'll see if you get February 100% correct, if you do, then April is at risk.
    27 Apr 2012, 01:59 PM Reply Like
  • dnpvd51
    , contributor
    Comments (2492) | Send Message
     
    "Over the past five to six months, lower natural gas prices have resulted in 7 Bcf/d of fuel switching from coal to natural gas for electrical generation in North America."

     

    Encana reports as above in their latest 6-K Many other companies report likewise. I think there is much fiction in their 6-K but the above might be true.
    27 Apr 2012, 11:57 AM Reply Like
  • Titan06
    , contributor
    Comments (301) | Send Message
     
    so with nat gas consumption from electricity producers at ~ 150 BCF/day currently and Encanna reports that low prices have added ~7BCF/d this correlates to a woppping ..............4.6666% jump in demand

     

    wooopdy doo !
    27 Apr 2012, 12:49 PM Reply Like
  • Mark Anthony
    , contributor
    Comments (3595) | Send Message
     
    Author’s reply » The 7 BCF a day did occur in some of the few past months. It was NOT due to price. It was due to temporary weather factor. More importantly, by April, 2012, the data shows it's COMPLETELY GONE already. Even though there is still 5 days missing in existing April data. I can already tell, with high confidence, that:

     

    Apr. 2012, 521.888 vs 525.956 a year ago, or DOWN -0.7735%

     

    See the data table.
    27 Apr 2012, 02:00 PM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    Mark, your projection of the data says that - we'll have to see if EIA confirms it with their data once they publish it.
    27 Apr 2012, 02:01 PM Reply Like
  • Marc Klein
    , contributor
    Comments (13) | Send Message
     
    Total nat gas consumption is running just under 70 BCF/day. Nat gas consumption for power generation from last jan and december (last 2 months of data) were approximately 20 BCF/day. 7 BCF/day switch from coal to nat gas, IF TRUE, is a major deal. That's a 10% increase in total demand and a 35% increase in electric utility demand for power generation.
    27 Apr 2012, 10:01 PM Reply Like
  • Marc Klein
    , contributor
    Comments (13) | Send Message
     
    Mark, I appreciate the level of detail in the above charts and respect your insights. And I am looking forward to your analysis on the inaccuracies of the weekly supply and demand data provided by Bentek and released by the EIA. I have noticed them as well. In fact, I wonder if the data for the week ending 2/22/12 has a misprint. They show a 49% year over year increase in nat gas usage for power generation, but a weekly decline of 16.8%. I wonder if that is a misprint and supposed to be a weekly increase of 16.8%. Looking at all the weeklies for the past 4 months this would be more consistent with the data. It would also help explain why your chart of weekly natural gas demand and supply based on weekly EIA demand and supply data , is inconsistent with the weekly storage data. Example, according to your chart, for the week ended 3/14, there should have been a slight injection but EIA reported a 64 BCF withdrawal. All weeks thereafter seem to show a larger withdrawal than actually occurred. Perhaps you will touch on this in your analysis of errors in the Bentek data. Thanks again for your analysis and insights.
    27 Apr 2012, 10:25 PM Reply Like
  • Mark Anthony
    , contributor
    Comments (3595) | Send Message
     
    Author’s reply » Paulo:

     

    It was NOT a projection. Projection is you make prediction to things that has NOT happened yet. April, 2012 is almost history, with a few days left to go. The folks who released those percentage figures have numbers they based on to caulculate the percentage change.

     

    What I did is simply try to deduce what those actual weekly numbers are, based on the percentages released and what other numbers we have been told. If the information provided is sufficient and accurate, and my mathematical method of deduction is correct, then the reusult must be correct as well.

     

    Projections have un-certainty as un-expected random events could happen. Deduction of numbers does not have un-certainty, it only has in-accuracy: calculation rounding errors, disprepancy between different statistical method, in-accuracy of some of the numbers provided. etc.

     

    My April number night not match to the precise last digit of the numbers accurately. But it will NOT be off much. Qualitatively the conclusion is already correct: This April does NOT seen a significant increase versus last April in terms of NG used in the power sector.
    28 Apr 2012, 10:08 AM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    It's a projection because you know that one of those sets of percentages is wrong, and you don't know which one - so we have to wait for EIA's future numbers and then see if they fit yours.

     

    Again, if you get February very close, then April should be very close too. It really needs to be VERY close, because if there's some deviation it might increase over time.
    28 Apr 2012, 10:15 AM Reply Like
  • Mark Anthony
    , contributor
    Comments (3595) | Send Message
     
    Author’s reply » Marc Klein:

     

    I noticed that unusual 49% but I do not doubt it. The week in the prior year was indeed much warmer and saw much smaller NG usage, thus it boost this current year's Y-o-Y increase to 50% while the usage is actually perfectly normal.

     

    Let's see the prior year same week's update, 2/24/2011:
    http://bit.ly/JsWVkf

     

    It indeed says it was much warmer and demand much smaller. NG draw down fell from 233 Bcf to just 81 Bcf.

     

    This once again shows just how MISLEADING it is to compare current week with the week one year ago,because you never know what kind of week you are comparing to. The 49% tells you nothing.

     

    Paulo saw that one 49% figure and he thought utilities have massively switched from coal to NG, when it was merely due to a much warmer, and much lower comparison figure a year early. This is how the percentages mislead people.
    28 Apr 2012, 10:22 AM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    There's a ton of weeks that show high demand.

     

    And besides, you just contradicted yourself and confirmed what I have been saying: that generally, warmer weather would have led to LOWER NG usage for power generation, however, this year we saw INCREASED usage because of the coal->ng substitution. Which means that the switch was even greater than it seems, because it is represented not only by the increased usage, but also by the inexistence of the decreased usage due to warm weather.
    28 Apr 2012, 10:30 AM Reply Like
  • Marc Klein
    , contributor
    Comments (13) | Send Message
     
    Mark,

     

    I am very excited for Monday's EIA release of February's monthly data and am very appreciative of you providing the above chart. I could find no errors in your chart showing us what the Bentek week over week data predicts for February monthly data. For february electricity generation, the versus last year BCF number should be approximately 618 (last year's 483BCF * 1.2825) vs your week over week number of 593. Of course I will be looking at all categories to see which method (vs last week, or vs last year) better explains what is happening in the natural gas market. Or whether the Bentek data is completely meaningless with regards to predicting EIA monthly data. Your chart will make that much easier for me decide, less work for me. By any chance do you know if the monthly EIA data is often revised? If so, are the revisions ever material?

     

    Thanks again.

     

    Marc
    28 Apr 2012, 03:10 PM Reply Like
  • Mike L
    , contributor
    Comments (169) | Send Message
     
    In addition to all the U.S. producers and insiders in 180 disagreement with this author on coal to gas switching via their earnings release and such,

     

    it looks like even in China, natural gas will be cutting in to coal's previously anticipated growth models for power generation, if this Russian pipeline deal goes through.

     

    *******

     

    April 28, 2012

     

    By Melissa Akin and Gleb Bryanski

     

    MOSCOW (Reuters) - Chinese energy officials signaled on Saturday they were keen to pursue a major gas deal with Russia despite a proliferation of alternative supplies, and said they had made a new proposal to overcome a price deadlock.

     

    Talks between China, the world's fastest growing energy consumer, and Russia, the world's largest conventional gas producer, fell apart late last year when China, frustrated by years of stalemate over price, increased its supplies from rival producer Turkmenistan.

     

    A senior energy official visiting Moscow as part of a delegation led by Chinese Vice Premier Li Keqiang, who is on track to succeed Premier Wen Jiabao, told reporters China had proposed a new model for cooperation on gas during the visit.

     

    "Now all that remains is the question of prices," said Liu Tienan, the head of China's National Energy Administration.

     

    He said Li had presented "a completely new model of development of cooperation ... in the natural gas sphere, and received a positive assessment from the Russian side."

     

    "We are interested in the companies of our two countries beginning discussions and consultations on this issue under the new model of cooperation."

     

    Earlier in the day, the chairman of China National Petroleum Corp. (CNPC), Jiang Jiemin, reiterated to an economic conference that most of the key points of a gas deal were agreed.

     

    "We have signed a buy-sell agreement with Gazprom," Jiang said, singling out the deal among successful ventures including a loans-for-oil deal, under which China receives 300,000 barrels per day via a dedicated pipeline, and joint oil exploration in Siberia.

     

    Under the terms of the gas deal, which was nearly finalized last year, Russia would sell up to 68 billion cubic meters of pipeline gas per year to China, more gas than it ships to any single European customer.
    28 Apr 2012, 04:56 PM Reply Like
  • vayu
    , contributor
    Comments (2) | Send Message
     
    Mark
    Your claim that there was no YoY growth in April 2012 is based on the weekly growth numbers. However, the same reports also show that YoY growth from power has been 25-42% every single week this year which means April would see a 30%+ YoY growth. Obviously, one of the two datasets is wrong. Any compelling reason for you to choose one over the other?
    29 Apr 2012, 05:35 AM Reply Like
  • Paulo Santos
    , contributor
    Comments (34872) | Send Message
     
    That's the reason we have to wait for the next EIA report with monthly numbers.
    29 Apr 2012, 09:01 AM Reply Like
  • Mark Anthony
    , contributor
    Comments (3595) | Send Message
     
    Author’s reply » My projection for Feb., 2012 power sector consumption of NG was 593 BCF. The EIA number came out to be 649 BCF. So I was off.

     

    Inspecting where I got it wrong. I noticed that the 2/23/2012 EIA weekly update looks problematic:
    http://bit.ly/Il7qnk

     

    It show the power demand was up +49% year-over-year, but some how it was down -16.8% from a week ago. That does not make sense.

     

    I checked the heating degree days, it was 203, a pretty normal cold winter week. The previous week was 175, warmer. So it makes no sense the temperature is colder but some how demand is weaker.

     

    Once I eliminate that -16.8% and replace it with 0%. The spread sheet then shows my Feb., 2012 prediction becomes 648 BCF.

     

    That's pretty spot on, missed by just 1 BCF!!!
    1 May 2012, 03:10 PM Reply Like
  • dnpvd51
    , contributor
    Comments (2492) | Send Message
     
    Injections are down year over year, and production is up year over year. So if there is no coal to gas switching how do you explain falling injections?
    14 May 2012, 12:06 PM Reply Like
  • Mark Anthony
    , contributor
    Comments (3595) | Send Message
     
    Author’s reply » dnpvd51:

     

    A picture explains things much better. See the chart here:

     

    http://seekingalpha.co...

     

    See the red line that stands for 2012 NG consumption in the power sector? It is trending down relative to last year, and is about to merge back to where it belongs. That tells you the bump was purely due to one time exceptional weather in the past few months.
    14 May 2012, 02:32 PM Reply Like
Full index of posts »
Latest Followers

StockTalks

  • Jim Rogers favors water investment! California is in a water crisis. Cardiz (CDZI) is up almost 50% today on news: http://tinyurl.com/qpddql
    Jun 5, 2009
  • China is still buying US T-Bonds and Yuan is pegged to dollar. WHY? Read my shocking discovery: http://tinyurl.com/rcuofx Enjoy!
    Jun 4, 2009
  • PEIX, a once hot ethanol play, goes Ch.11 today, as expected. Recent unusual price moves must lured speculators in. Good lesson learned.
    May 18, 2009
More »

Latest Comments


Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.