“Prophesy is a good line of business, but it is full of risks.” Mark Twain
It has been quite some time since I have posted an article.
The end of the year is a good time to reflect on past prognostications. I do have a list of publicly made prognostications. Of course, my most famous prophecy was the collapse of the oil markets in 1998. I was noted in several publications including front page USA Today. Since then, given my move into the real world from consulting world, my prophecies have been privy to the companies at which I’ve worked. However, you can find some public appearances at conferences where my prognostication skills can be bench marked to reality, I have dug up two from the past – one in 2011 and another while I was running my consultancy in 2015. Both, coincidentally, were at a Platts conference. I seem to do quite well at Platts conferences – perhaps I should be presenting more often there. Last, but not least, the latest news of auto manufacturers layoffs and cutting models made me go back to one of my previous blogs I posted.
The last public statement was made at the Platts Refinery Conference back in 2015. Right after that, I was implementing PMA at EDF Trading. At the conference, I was not given much attention as many of my ideas did no support the on-going trend and sentiment. In addition, I hadn’t spent a lot of time back in the refinery industry – but to be honest, its really not as complicated as power – for me it was easy to pick back up where I left. I rebuilt supply/demand models for each commodity. Reviewed and traced each demand and supply sector and examine key inflection points. I also followed up with respective experts/leaders in each product and had wonderful conversations with them. An expert with passion can’t stop talking :)! Please click on this link for the 2015 Platts Refinery Conference Presentation. I started my presentation noting how all the experts in the industry over the past decades have gotten major trends wrong and I wasn’t excluding myself (e.g. lighter crude slate – peak oil etc…). Many in the rooms did not like to hear this. However, if you ever get a consultant/expert to not review and face the facts of the past and admit their errors – run – run as fast as you can. An over confident consultant/expert can cost you a lot more than their fee. I discussed the current landscape with all the new oil and gas production and how ethane will need to find a new home. Currently ethane exports are almost at my 2024 foretasted export level! Also noted was the possible Naphtha issue and the two chemical feedstock would inevitably collide in the market. All of it came true, and the extent was even more than the tempered forecast I presented. I am a realist in the sense there is no reward to forecasting way above consensus – just being away from consensus is enough to get the point. I had an inclination that the situation could be much more dire than I presented, but I was already going against the grain so I tempered the expectations in the presentation. Creating scenarios also allowed the visuals of the direction of the market – note the high case.
Going even further back is a presentation I gave at Platts Coal Conference in 2011. I was in a similar situation where I had to relay a bad message to the room – and once again, many did not want to hear it. To view the presentation click on the following link – Platts Coal Refining Conference in March 2011. I noted my calculations of coal retirements made when I was Managing Director Strategic Planning for AEP – which was available in the public documents in one of the AEP Analysts meetings presentation. Essentially I was telling the audience that almost 50% of their demand might go away and that exporting would not solve their financial situations. Of course, no one in the coal conference wanted to heed my warning. The KOL etf had already dropped from highs of 50 in the year to 30’s in October 2011. I heard from some that this David person had no idea what he was talking about – because I was a power guy not a coal guy. Hmmm… when most of your demand is from the power sector shouldn’t that be the person you should be listening to? Anyways, the KOL etf is now 12 and we have seen some major bankruptcies in the industry. Export coal is up in volumes but just not even close to drop of volumes from US coal generation. The surprising reflection to note is how well the rail sector has done – perhaps replacing coal with oil and continued Chinese imports – but at some point that will likely end, too.
The latest news of auto manufactures laying off and cutting models made me think of an article I posted back in Feb 2015 which I reviewed the BP Energy Outlook. If only the auto guys took heed to my warnings – “…eventually with the falling prices and the improved efficiency improvement the auto manufacturer can produce an SUV with mass appeal and size that can go 0 to 60 in few seconds yet offer 25-40 mpg. Auto manufacturers who ignore this trend will be left in the dust as was seen last time SUV sales outsold compact vehicles. The move to this larger and faster car will swallow the small vehicles leading to overall growth in oil demand while maintaining the CAFE standards.” US total petroleum demand almost re-hit its monthly peak demand set in August 2005. Overall US petroleum demand has been in an upward projection since bottoming after the financial crisis.
There are many out there denying the ability of human beings to predict/understand the future. Books like Undoing Project by Micheal Lewis and the conclusions that Israeli psychologists Daniel Kahneman and Amos Tversky demonstrated that humans have a breakdown in their psyche creating poor decisions and inefficiencies in the market. They did not examined learned/experienced forecasters and also did not reach out to successful prognosticators. Much of their tests involve normal sample people, not experienced and trained-to-forecast individuals. We are also seeing the rise of AI and neural networks to take the human psyche out of the equation for decision making. However, it is the human that can relate to the human that is making decisions, even if the human is using AI. So far, I have yet to see an AI be able to go beyond short term analysis given its learning sample is always based on the past and the multiple forward inputs still require some creativity and art for which only an experienced person can create. I good benchmark for when AI is ready is when AI starts making art work so profound. At that point, I would say human prognosticators could likely be looking for another career, but until then – For my fellow aspiring prognosticators I leave you with these tips in order to be a better forecaster.
- One. Get a real job in the real industry and eventually in the planning and strategy group. Being a consultant your entire career leaves out the feedback loop of reality and you end up in this strange world of advice giving and not knowing how it really matters to the company and the multiple lives you could impact. This is also a note to companies hiring consultants – get one that has been in the industry in a position of planning/forecasting. They will understand your position best and know the bigger picture of what an outlook/position could mean to you and your company.
- Two. Have a fundamental foundation – examine as deeply as you can the drivers of supply and demand e.g. what are the economics parameters that drive more supply? When and how much does demand respond to price? What alternatives are there? etc.. The devil is in the details – get your hands dirty and get into the details. This is where IF you have passion, it will shine. IF you want to be the best, this is where sacrifices are made. Work longer – sleep and dream the problems – let nothing get in your way – if you really want to understand the market your forecasting. If you don’t, I promise you someone else will, and you won’t be getting it right consistently. Forget all those that say you must sleep so much and rest so much – I believe, when your in the midst of an issue, you must engross yourself and get your rest and leisure later. Nothing worthwhile that can be claimed as a great achievement comes without sacrifices. My first professional boss at Purvin & Gertz, Ken Miller, taught me about having passion for your work. His life was his work. He would carry the largest suitcase full of reading materials everyday. I have never stopped reading as much as I can because of him. He passed away this year, and may he rest in peace – but, knowing Ken, his rest would be reviewing refinery economics. Everyone should have such a boss with so much passion. There is so much that I owe him for showing me the way.
- Three. Be humble – know that you don’t know everything nor do you have to. In the mid 2000’s, I knew gas couldn’t just rise to the teens without something in between. Did I know it was shale at the time? No. But, I did know the principle of greed and human ingenuity. My premise to pull the prices down from the every growing forward curve came from the underlying thought of greed and human ingenuity, and I used LNG as my placeholder. In my calculations, at the very most, prices would be around $6-7/mmbtu. In the end, we see the development of shale was the ultimate greed and human ingenuity mechanism pulling gas prices back down even farther than I foresaw at that time. However, my premise was still right. It was not ever growing as the forward curve/market thought, and I didn’t have to know everything to save the company billions of dollars from jumping into more coal assets and not buying distress gas assets that later sold for multiple times on the dollar. It is so important to reach an understanding at some point in your career that no matter how much modeling and research is done, the world will unfold not even close to many experts predictions. With that humbleness you will open your mind and begin to, more accurately, prognosticate. You will realize and be more open to multiple outcome probabilities. Having ranges of outcomes does not make you weak, but makes you cognizant of the multiple dependencies that are out of your control e.g. weather. However, you will still need to have a base forecast to discuss where things are likely headed as long as all the dependencies likely head your direction.
Over the years, I have come to the realization that humble, but passionate, consultants are the best consultants, as they have been weathered and understand some of the unpredictable nature of the past and very much willing to put in the effort required to maintain expert status. They are confident enough to change their forecast before it is too late and also be brave enough to forecast against the trend.
May you have a prosperous 2019 – and stay thirsty to knowledge!
David K. Bellman
Your most humble, super passionate, and grateful but very lucky prognosticator of the energy market!
Clean Power Plan will Double Coal Retirements Unequally
The article I wrote in the October 2014 issue of Public Utilities Fortnightly “EPA Clean Power Plan – An Unequal Burden” (Click for free copy) has been cited in numerous responses by commissions around the country. In this article, I will clarify the intent and introduce another layer of analysis.
The analysis was not funded by any entity. The development of the analysis came from pure intellectual curiosity and the enthusiasm of solving a puzzle. All Energy Consulting has not received any payment or business as a result of the analysis done on EPA Clean Power Plan. The article is not intended to argue the merits of global warming. The article is also not intended to absolutely quantify the cost of the plan. The main intention of the report was to demonstrate that the cost of the plan will vary significantly among the states. In order to do this, as noted in the article, we used the implicit EPA calculations for Blocks 3 & 4. In addition, we used our highly sophisticated and very well calibrated power modeling platform – Power Market Analysis (PMA). PMA is used to quantify risk in the futures power markets for hedge funds to end-users. PMA was used to calculate an impact of Block 2. Block 2 per EPA involves the re-dispatching of the system with gas generation over taking coal generation. PMA uses the software AuroraXMP by EPIS which allows us to input our knowledge in a relative easy manner and produce numerous runs in a short time span. We had to produce hourly dispatching results for the entire N. America with at least 15 plus simulations to produce the analysis. For the analysis, we re-dispatched the system using state by state carbon prices in order to achieve the CO2 emission rates targeted by EPA for Block 2 – see process flow figure below.
There are limitations with this analysis for use outside the intended use of demonstrating large economic disparity among states as result of the Clean Power Plan. Given our goal of showing large differences in state economic impacts, we did not do a full cost analysis of the plan. EPA simple analysis did not do this either. They just assumed the substitution of generation from coal to gas would be sufficient given the perceived underutilized capacity factors of gas plants. However, a full cost analysis of the plan would involve an iterative approach to retire units no longer economically sustainable. As the CO2 costs rise, many of the existing units no longer produce enough power to be economically supported given their fixed cost of operation. This leads to retirements in the system. There will be stranded cost issues with early retirements of plants. Our initial analysis did not go that far. The cost derived from CO2 prices to drive those retirements represents a conservative cost that will occur in order to substitute with new technology. In many cases, the cost will even be higher than the CO2 impacts of driving units – mainly coal – out of the system given replacement power needs an economic hurdle to get a project developed. We can certainly do this analysis at All Energy Consulting, but we were limited by time and the need to pay the bills. Given this analysis was done free of charge, and our intellectual goal was to demonstrate large state disparity, we were able to achieve this without a retirement sequence analysis.
We were given more time recently and did do the next sequence of the analysis –once again for pure intellectual curiosity – no payment. Taking the output from our 100% Block 3 & 4 case from the article, we did a unit by unit analysis and identified the units that are no longer economically viable vs. the no CO2 case. This amounted to nearly 60 GW of generating units with almost 90% being coal – see map below. This is on top of the already planned retirements. In our base simulation, we have nearly 60 GW of coal plants retired from 2010 to 2016. Gas units are under economic pressure if the plant is located in a state with high carbon prices relative to the surrounding states. There will be gas plant casualties as a result of plant location as electricity flows do not know state boundaries.
We then retired those units and re-dispatched the system and went through the process as noted above to converge upon a CO2 rate that would meet EPA rate limits. This took almost 20 simulations given the need to also re-compute the retirement economics as we lowered carbon prices.
As expected CO2 prices do fall down relative to the previous results as coal units disappear from the system.
Reliability concerns are likely, but can be managed with transmission investments. These costs are not trivial and will show up in retail rates not wholesale prices. Retirements will be balanced by the large expansion of solar, wind, and energy efficiency programs and likely new gas generation from neighboring states. Wholesale power prices in this case are lower than in the previous case given the lower CO2 prices, but for Texas as 24% of the retirements came from there – see below.
Texas is expected to get a large dose of wind generation so the wholesale prices are likely to come down and potentially driving even more retirements – see below.
Obviously a crucial element was the outlook of the spread between gas and coal prices. Staying with the Fortnightly analysis we used the forward curve projections of 2016 in June 2014. Henry hub averaged $4.26 with CAPP coal prices at $65/ton. Both those benchmarks are much lower now – 25% lower for Henry and 20% for CAPP. Lower spreads likely means more coal retirements.
The next step of the analysis is to create a build profile to meet the requirements in Block 3 & 4 and re-disptach it within the system. This will have significant cost plus significant dispatch implications. One of the areas, we are reviewing is the large solar penetration in the west which will likely drive peak power prices negative in the shoulder months. All Energy Consulting can do this analysis, but we are looking for interested parties to fund this stage of the analysis. The result will be a comprehensive cost of the plan. In addition, we will be able to address some of the impacts of such large renewable penetration. Those interested in funding the analysis will get customized views of the results.
If you enjoy this type of analysis, please do consider supporting All Energy Consulting by contacting us for your energy consulting needs. We are able to offer these types of analysis because of the support we get from our clients. All Energy Consulting offers a pragmatic but enlightening approach to analyzing the energy markets. Our focus will always be for your success.
Your Enthused Energy Analyst,
David
Contact Information:
David K. Bellman
Founder/Principal
All Energy Consulting LLC
[email protected]
614-356-0484
Background David Bellman was the former Managing Director Strategic Planning at American Electric Power (AEP). He also worked as a consultant in Deloitte Consulting and Purvin & Gertz – now part of IHS. All Energy Consulting was formed in 2011 focused on energy analytics in order to add insights to the energy markets.
Coal to Gas Switching – 2012 Déjà vu?
This report available in PDF Form.
The amount of gas and coal burns in the power markets go beyond simple economics when the price spread between gas and coal are very narrow as observed in 2012 – see figure below. Those attempting to use supply stacks or historical calculations will likely see incorrect numbers. When the price spread is high, those methods will work quite well, as long as the markets do not observe large retirements or new builds in the power fleet.
To demonstrate the issue with low gas to coal price spread, we present our calibration results of fuel burns over the last few years. Please note the 2012 discrepancy. One could conclude our models are fatally flawed.
However, the goal of PMA is to understand the markets and quantify the risk in the market place. The calibration run represents the default runs that we run forward in time in our BASE case. The key inputs, which drive the BASE case, are weather, economy, and commodity price. Two large areas not demonstrated in the base inputs are outages and utility operations. However, these are simulated in our risk runs that run daily with the BASE runs. The default outage profiles in the BASE case were developed based on a unit by unit investigation over a 5 year period. For the 2014 deviation, much of this can be attributed to outages which are captured in our risk cases. The winter deviation was discussed in our winter assessment. In addition, greater maintenance outage occurred in the shoulder months in 2014 as units have been installing FGD, SCR, and baghouses for the EPA regulations. The relevant issue for 2015 summer is how coal generators plan to operate and dispatch their plant. In the BASE case, utility operations are assumed to be market driven. However, in 2012, this is not what happened for several coal units.
To prove this, we re-ran the BASE case with modifying the market behavior of several coal units. Now the figure of coal consumption matches quite well in 2012.
The issue in 2012 which resulted in the need to change the operation mode of coal units was a result of the coal purchased in 2011. The futures market curve in 2011 for 2012 showed a much larger spread in gas and coal prices and also a robust power market price. This resulted in a much larger purchase of coal than required in the market place in 2012. This caused coal inventories to be very high- see figure below. This dynamic of coal purchasing was a result in the shift in coal and power markets – see our discussion on Coal Market Changes below.
Many plants cited hazardous conditions of stock piles resulting in the need to use the coal regardless of power plant economics. If this behavior did not occur, the trajectory of coal inventories would have gone much higher.
The reason not to adopt this method in the BASE case is the issue with power prices and the principle of market behavior being rationale over time. As expected, if the coal generators ran uneconomically, they would distort the price – see figure below. There is a Heisenberg Uncertainty Principle in power modeling. The volumes and price of the market cannot be discovered simultaneously in the same level of precision. As soon as you get your volumes correct your prices will be off. There is a give and take between those two. Given the model was developed for trading power prices and understanding the power market price risk – we prefer to be more accurate on prices than fuel burns. However we can adjust the settings for our clients need.
Most utilities still use the futures market as their guideline to purchasing. They should run multiple scenarios and situations using a platform like PMA to really understand the meaning of the futures market. The futures market clearly weighs the risk of near term events more than the actual possibilities. PMA can correctly quantify this risk, so you can make an educated decision on your fuel purchase. Not only can we quantify economy, weather, and commodity price risk, but we can quantify the other facets of the market place such as utility operations presented here.
Our 2015 summer assessment (May-Sept) shows if we applied the same utility modifications done in 2012 we can see an increase of coal consumption of 14% from our base case. The likelihood of this is less given the current coal stocks levels are lower plus the change in the market place as several coal units are de-regulated. Before, many of the coal plants could get fuel recovery from their un-economic dispatch – see more explanation below in our discussion Coal Market Changes. This is less likely than in 2012, therefore there is less incentive to run uneconomically.
Contact us to get a range of coal and gas consumption in the power sector plus power prices. The current futures market is showing a very tight spread between gas and coal. This may reduce price volatility, but it will make up for it in coal and gas volume volatility.
Also, consider creating your own set of conditions and parameters to run through PMA to get a much greater insight in the market space. PMA enables you to use your knowledge applied to our market knowledge.
Please consider our services – help us help you succeed.
Your Ever Evolving Energy Analyst,
David
David K. Bellman
Founder/Principal
All Energy Consulting LLC- “Adding insights to the energy markets for your success.”
614-356-0484
[email protected]
@AECDKB
Sign Up to AEC Free Energy Market Insights Newsletter
Changing Coal Markets
An appreciation of what was occurring in the coal and power market pre 2012 is important to understand the current coal dynamics in the power markets. In the past, coal contracts were developed over long periods of time (some over decades). Overall commodities prices were rising worldwide, including coal. Many coal producers no longer wanted to have long-term contracts in favor of shorter term allowing them to participate in this rising environment. (Oh how they wish they didn’t think that way now – greed gets the best of us.) Coal generators felt the squeeze, particularly, if you had to purchase spot coal. The price of spot coal was very high and would be typically $10+/ton more than contract coal. During the height of the export markets, there were possibilities of not being to obtain coal at any price. Therefore, there was much incentive to lock in most, if not all, of your expected coal purchase ahead of time, even if it was for just a year.
Also contributing was the deregulation aspects of the power market. Before, a utility could plan and make a decision to produce power at fixed price given a fixed price contract for coal. The regulated utility had a mechanism to fully recover all their fuel cost. However, with deregulation and the developed option to obtain power from a market versus your planned generation, this caused a disconnect in the way things were done and how they can be done going forward. This new construct put several utilities in a position to not want to be bounded by too long of contracts either.
Coal operators and coal generators need to come up with new ways to align themselves. I believe coal contracts need to be structured on spark/dark spreads allowing protection from low gas prices for the generators to ease inventory concerns and at the same time giving coal producers the upsides during peak time periods. If you are interested in this process, All Energy Consulting will be happy to demonstrate and work with both the coal generator and coal producer to structure a win-win situation.
Whitepaper – Key Pieces to Modeling Power Effectively
As noted in the beginning of the year for 2015, we will work on producing a clear message to our readers, clients, and prospective clients about our commitments and value proposition. We are committed to adding insights to energy markets for your success – today and for years to come.
As part of the commitment, we begin with knowledge sharing by introducing a whitepaper on power modeling. Understanding the key pieces of modeling power will allow you to begin your endeavor in this challenging but very rewarding effort. This whitepaper will outline the areas of focus and also make a few high level points. We hope this will help you in your efforts, and as always, we hope you consider All Energy Consulting as an option for assisting in implementation or potentially using our PMA platform as an introduction into your power modeling efforts.
Please Click Here to Download the Whitepaper – Key Pieces to Modeling Power Effectively
AEC realizes you have many options in the markets when it comes to energy consultants. I want to state now – no other consulting company will be more committed to sharing knowledge – not just information – for your success than All Energy Consulting. We are not hesitant to empower you and your team to the point where our consulting role may be limited in the future – we treat this as a success.
Your Very Willing to Share Energy Analyst,
David
David K. Bellman
Founder/Principal
All Energy Consulting LLC- “Adding insights to the energy markets for your success.”
614-356-0484
[email protected]
@AECDKB
Year of Model Building 2014
2014 Modeling Efforts
At All Energy Consulting, we specialize in modeling the energy markets. To give you a sense of our capabilities, below are the list of models built in 2014.
Reviewing the list, it reminds me of the saying :
“People who love what they do wear themselves down doing it, they even forget to wash or eat….When they’re really possessed by what they do, they’d rather stop eating and sleeping than give up practicing their arts.” Marcus Aurelius, Meditations
Please do consider All Energy Consulting for your energy modeling needs. I truly enjoy what I do and it will show in my work and commitment to you – Thanks for 2014.
David
[email protected]
614-356-0484
Oil & Gas
-
Built a USGC refinery model in various configurations – Hydroskimming, Cracking, and Coking
-
Discounted Cash Flow Model for Refinery Acquisition and Refurbishment
-
USGC and Caribbean petroleum pricing model
-
Modeled the US Natural Gas Deliveries to Electric Power Consumers by State
-
Modeled and Calculated the discount value for Eagle Ford Condensate
-
Built World Supply/Demand Balance Model for Crude Oil and Petroleum Products
Power
-
Assisted Platte River Power Authority (PRPA) in deployment (setup, build, and operate) of a power model used for their Integrated Resource Plan
-
Several Discounted Cash Flow Model for Power Generation Asset – 800 MW CC to 1MW Reciprocating Engine including the associated power modeling work
-
Worked with the University of Texas Center for Energy Economic on modeling and publishing papers regarding ERCOT
-
Modeled EPA Clean Power Plan and published paper in Fortnightly
-
Modeled and produced a risk analysis for a set of generating assets
-
Long-term Power Modeling along with integrating GCPM gas pipeline model
-
Built and Operate Power Market Analysis (PMA) Platform – includes integrated trade screeners to integrated natural gas storage models.
-
Built an interactive load model for 118 load zones representing N. America
-
Built a coal pricing model to deliver and price coal to all 1000+ coal units in N. America
-
Discounted Cash Flow Model for an Integrated Desalination Plant with Associated Power Generation and Recycle Facility for Waste Water from Fracking
Models are only as good as the inputs and the ability to decipher the outputs to business solutions. More information on modeling and other services can be found on our website.
Best of Market Insights 2014
To my readers, clients, and prospective clients,
I appreciate your feedback throughout 2014. The year 2014 will be remembered as the year of discovery for All Energy Consulting (AEC). An identity for AEC had to be discovered – better late than never. I know I had to work more on delivering my value proposition to you. I have spent so much time on developing products/services plus analyzing markets I lost sight of delivering this message. You will see an enlightened AEC moving into 2015. Our value proposition to my readers and prospects will be clear. I want to share the current thoughts on this in our Branding Positioning Statement:
“For those who need forecasted energy commodity prices, we are experienced market analysts who assist in the navigation of uncertain energy markets with a proven process and methodology, and a collaborative approach in consulting that yields clarity, transparency and empowers decision-making.”
You will see a change in 2015 with a clear direction to support the statement above.
Below are the best of market insights in 2014 rated based on web statistics – page views and downloads.
Your Very Grateful More Focused Energy Analyst – Happy New Years!,
David
614-356-0484
Best of Market Insights 2014:
– With the most page views (60K+) –Peak Energy – Are we there in the US? This is very surprising. Perhaps this is a result of peak oil conspirators or just an error in the web stats. The article uncovers the decoupling of the economy with load in several areas in the US.
– Excellent Returns Were Produced From PMA Summer Model Predictions This article highlights the performance of the generic PMA model for the summer. PMA correctly highlighted the hype from Polar Vortex bleed into the summer forwards. Also should read theBest Winter Trade which yielded 30%.
– Ready for March Gas Demand? Duke’s Merchant Coal Plant Value This article demonstrated PMA flexibility to not only help create risk-adjusted forward curves but also calculate natural gas demand and do specific power plant analysis. We also highlighted our agility with the ability to deliver results within 24 hours.
– US Refining Margins Outlook Sept. 2012 – Wow. An oldie had nearly 20K page views. This article highlighted that condensate production was going to change the markets and that US refining margins will be robust for years to come. Well it was an UNDERSTATEMENT – even though I was probably one of the few analyst mentioning the significance of condensate in 2012 the production of condensate blew past my old outlook. US Refining margins were and will be robust! See 2014 US Refining Outlook
– Most downloaded (403 downloads) – Summer of 2014 Analysis – I threw the kitchen sink and all into this analysis. If you took the time to really digest all of the information in the report you would have been prepared for what transpired in summer 2014.
