Electricity Trading

I will be interning for a energy trading firm, and I was wondering if you guys could help me out with some of the mechanics of trading. I understand basic power grid operations and how the financial markets work, but not too sure about the specific transactions for energy traders.

Please correct me if I'm wrong anywhere in this posting. From what I understand, since electricity is consumed as soon as it is generated and cannot be stored, there is a constant goal of balancing electricity (this is more of the power operations side).

Also, there are two types of trading in the energy market, real-time and day-ahead forward trading. From what I have gathered, day-ahead trading is the most active, and real-time trading plays more of a balancing role.

The firm I will be interning for is a prop shop, and I will be trading on the firm's capital. These are some of the major questions I have, any one of your guys answers were be deeply appreciated, as I have been looking for these answers for a very long time. They all deal with the mechanics of trading:

-In the day-ahead market, where do traders purchase electricity contracts from? Is it from a RTO/ISO, is this analogous to a stock exchange?

-What exactly do day-ahead traders do? Do they purchase forward contracts one day ahead of time, and lock in a set price for a fixed amount of electricity, and depending on the amount actually produced by generators, they either make or lose money?

-Day-ahead traders don't sell to anyone in this market, correct? They just submit bids for contracts, betting on the amount of electricity produced?

-What is the difference between an ISO/RTO and an exchange?

-In very general terms, what do traders provide for the electricity market?

Thanks in advance for your help, it is greatly appreciated

 

DA vs RT in terms of liquidity is a toss up. RT is most liquid in MISO/PJM/ERCOT. The Nepool and West markets are mostly traded DA. In terms of size traded, PJM dwarfs most of the other markets in daily liquidity.

-In the day-ahead market, where do traders purchase electricity contracts from? Is it from a RTO/ISO, is this analogous to a stock exchange?

Other traders on ICE.

-What exactly do day-ahead traders do? Do they purchase forward contracts one day ahead of time, and lock in a set price for a fixed amount of electricity, and depending on the amount actually produced by generators, they either make or lose money?

Speculate on DA clears or RT clears basis their fundamental knowledge. If you think weather is bullish for power tomorrow, like a major heatwave in the summer, you would want to be long RT power.

-Day-ahead traders don't sell to anyone in this market, correct? They just submit bids for contracts, betting on the amount of electricity produced?

No. They buy and sell with other trading houses and utilities. They make markets for utilities that need to hedge their positions. If you're an IPP in PJM, you don't have any load to "put" your power to. So you might sell blocks of term power?

-What is the difference between an ISO/RTO and an exchange?

ISO/RTO is the physical power market that clears power on a daily basis. An exchange is where the contracts that trading companies and utilities are cleared.

-In very general terms, what do traders provide for the electricity market?

Liquidity.

Thanks in advance for your help, it is greatly appreciated

 
 
Best Response

Hey Valuestrats920, I’ve worked at/within PJM for 6 years.

There are three main products that are traded in the PJM Market (excluding other exchanges such as ICE/NYMEX). They are virtual bids (increment offers (INC)/decrement bids (DEC)), Financial Transmission Rights, and Up-To-Congestion. I will focus primarily on virtual bids with this post, but please let me know if you would like to know details about the others.

Virtual Bids bet against the spread between the Day-Ahead (DA) and Real-Time (RT) price (LMP) for each pricing node (over 9,000) and hour. For example, a decrement bid for the Western Hub at Hour Ending (HE) 16 will yield a revenue of (RT LMP-DA LMP)*MW. So, if you were to bid and clear 100 MW for HE16, and the DA LMP is $40 and the RT LMP is $100, your revenue is $6,000. Conversely, if the DA LMP is $100 and the RT LMP is $40, your revenue is -$6,000. With a DEC, you are betting on the RT LMP being higher than the DA LMP, or in simpler terms, you are buying electricity in the DA Market, and selling it back in the RT Market. If you place increment offers, you are selling electricity in the DA Market, and buying back in the RT Market. Hence, you want the DA LMP to be higher than the RT LMP.

A decrement bid is a purely financial instrument, but is interpreted as a load in the dispatch model of the Day-Ahead Market. An increment offer resembles a generator. When the market clears in the Day-Ahead, there is no distinction between a virtual bid and an actual physical product. In other words, if a hedge fund were to place an increment offer for 100 MW @ $50/MWh at XYZ node, and an actual power plant also places it’s generation bid for 100 MW @ $60/MWh at XYZ node, the increment offer clears instead of the actual physical power plant.

There are three main functions of virtual bids. One; it provides liquidity to the market. It helps to converge prices between the DA and RT (they are called convergence bids in California). If the spread between the DA and RT LMP at a particular node is large over a long period of time, traders will enter the market to capitalize on the arbitrage, and will eventually converge the prices. For example, if the RT LMP at XYZ node is always greater than the DA LMP, people will start to place decrement bids at the node (demand), and hence increase the DA LMP. The DA LMP will eventually start to converge upon the RT price. This provides better stability in prices for the actual physical participants. Two; it helps control market power. As explained in the example in the previous paragraph, an increment offer can compete with actual generators. So, if a generator tries to exercise its market power by offering power at $999/MWh, an increment offer can come in and offer a lower price. This helps to keep physical owners of raising prices. Three: it is used as a hedging mechanism for participants with physical deliveries. Say a utility has 5,000 MW of generation. They have to offer this full amount in the DA Market in accordance to the Tariff. They will receive the DA LMP * MW. Let’s say the DA LMP is $50/MWh. This is a revenue of $250,000. If the utility does not deliver the full 5,000 MW during the operating day, they are subject to buying the shortfall at the RT LMP. So, let’s say the utility only produces 4,900 MW in the RT, and the LMP is $100/MWh. They need to pay PJM (4900 MW-5000 MW)$100= $10,000. Their total revenue is $240,000. If they know their generators typically only produce 98% of their scheduled MW, they can place a decrement bid, which will generate revenue if the RT LMP is greater than the DA LMP. Using the same example with a decrement bid of 200 MW, the bid would yield ($100 RT LMP - $50 DA LMP)200 = $10,000, which exactly offsets the loss by not producing enough physical power. The same strategy holds true for utilities buying power, but in the opposite direction.

As far as what causes the differences between the DA and RT… there are a ton. The most basic reason is the difference in forecasted supply/demand (DA) vs actual supply/demand (RT). If PJM forecasts there will be 150,000 MW of load in the DA, but there is only 140,000 in the RT, the price will be (generally) lower in the RT. Another major reason is power plant/transmission line outages. If a power plant suddenly shuts off in the middle of the day, it will cause RT prices to spike drastically, as this outage was not scheduled in the DA. The physical nature of the transmission lines is also a huge factor. If a transmission line becomes constrained in the RT that was not modeled in the DA, it will cause RT prices to be higher, especially if in a very constrained area (load pocket). Imports and exports between the various ISO’s cause constraints along the “seams” of the grids. Fuel (mainly natural gas) on the different pipelines also plays a significant role in prices.

As I said, there are many, many reasons that cause these differences, and while general trends can be modeled with statistics, knowing the fundamental operations of the grid will allow you to predict the outcomes much better. Since I worked at PJM for many years, I was able to learn where all of the generators are located, what prices they offer, their physical characteristics, and their affects on the grid… what the limits on the transmission lines are… what the natgas pipeline prices can indicate… as well as many other fundamental reasons for how the prices are being set. An understanding of the actual pricing algorithims also helps. You could be the greatest mathematician in the world, but unless you have an understanding of what is fundamentally driving the prices, you will not do well trading in electricity markets. It is not like other markets, in which prices are more influenced by financial traders. With electricity, the RT price is always set by the physical operation of the market, which can not be influenced by financial traders.

Hope this helps a little, and please let me know if I can explain more. I could write for hours about this. If you don’t mind me asking, where are you interning? Name of firm is not necessary, but how about location, size of firm, and type (hedge fund/energy prop shop/utility/etc…)? What is your background?

 
thompsonpsu:
It is not like other markets, in which prices are more influenced by financial traders. With electricity, the RT price is always set by the physical operation of the market, which can not be influenced by financial traders.

Good discussion...but I'm not sure I agree with this statement.

Check out page 13, figure 2-17 here - http://www.monitoringanalytics.com/reports/PJM_State_of_the_Market/2012…

Virtual activity was marginal in the DA 96% of the time. Say someone DEC bids 10GW of load in a market with 5GW of load, a la Kirkpatrick in NYISO. This would over-commit in the DA, and subsequently RT prices suck. Your statement just completely ignores the futures/swaps side of the business, which, in my opinion, drives much more market activity than simple inc/dec/up-to profiteering.

 
 
thompsonpsu:
Hey Valuestrats920, I’ve worked at/within PJM for 6 years.

There are three main products that are traded in the PJM Market (excluding other exchanges such as ICE/NYMEX). They are virtual bids (increment offers (INC)/decrement bids (DEC)), Financial Transmission Rights, and Up-To-Congestion. I will focus primarily on virtual bids with this post, but please let me know if you would like to know details about the others.

Virtual Bids bet against the spread between the Day-Ahead (DA) and Real-Time (RT) price (LMP) for each pricing node (over 9,000) and hour. For example, a decrement bid for the Western Hub at Hour Ending (HE) 16 will yield a revenue of (RT LMP-DA LMP)*MW. So, if you were to bid and clear 100 MW for HE16, and the DA LMP is $40 and the RT LMP is $100, your revenue is $6,000. Conversely, if the DA LMP is $100 and the RT LMP is $40, your revenue is -$6,000. With a DEC, you are betting on the RT LMP being higher than the DA LMP, or in simpler terms, you are buying electricity in the DA Market, and selling it back in the RT Market. If you place increment offers, you are selling electricity in the DA Market, and buying back in the RT Market. Hence, you want the DA LMP to be higher than the RT LMP.

A decrement bid is a purely financial instrument, but is interpreted as a load in the dispatch model of the Day-Ahead Market. An increment offer resembles a generator. When the market clears in the Day-Ahead, there is no distinction between a virtual bid and an actual physical product. In other words, if a hedge fund were to place an increment offer for 100 MW @ $50/MWh at XYZ node, and an actual power plant also places it’s generation bid for 100 MW @ $60/MWh at XYZ node, the increment offer clears instead of the actual physical power plant.

There are three main functions of virtual bids. One; it provides liquidity to the market. It helps to converge prices between the DA and RT (they are called convergence bids in California). If the spread between the DA and RT LMP at a particular node is large over a long period of time, traders will enter the market to capitalize on the arbitrage, and will eventually converge the prices. For example, if the RT LMP at XYZ node is always greater than the DA LMP, people will start to place decrement bids at the node (demand), and hence increase the DA LMP. The DA LMP will eventually start to converge upon the RT price. This provides better stability in prices for the actual physical participants. Two; it helps control market power. As explained in the example in the previous paragraph, an increment offer can compete with actual generators. So, if a generator tries to exercise its market power by offering power at $999/MWh, an increment offer can come in and offer a lower price. This helps to keep physical owners of raising prices. Three: it is used as a hedging mechanism for participants with physical deliveries. Say a utility has 5,000 MW of generation. They have to offer this full amount in the DA Market in accordance to the Tariff. They will receive the DA LMP * MW. Let’s say the DA LMP is $50/MWh. This is a revenue of $250,000. If the utility does not deliver the full 5,000 MW during the operating day, they are subject to buying the shortfall at the RT LMP. So, let’s say the utility only produces 4,900 MW in the RT, and the LMP is $100/MWh. They need to pay PJM (4900 MW-5000 MW)$100= $10,000. Their total revenue is $240,000. If they know their generators typically only produce 98% of their scheduled MW, they can place a decrement bid, which will generate revenue if the RT LMP is greater than the DA LMP. Using the same example with a decrement bid of 200 MW, the bid would yield ($100 RT LMP - $50 DA LMP)200 = $10,000, which exactly offsets the loss by not producing enough physical power. The same strategy holds true for utilities buying power, but in the opposite direction.

As far as what causes the differences between the DA and RT… there are a ton. The most basic reason is the difference in forecasted supply/demand (DA) vs actual supply/demand (RT). If PJM forecasts there will be 150,000 MW of load in the DA, but there is only 140,000 in the RT, the price will be (generally) lower in the RT. Another major reason is power plant/transmission line outages. If a power plant suddenly shuts off in the middle of the day, it will cause RT prices to spike drastically, as this outage was not scheduled in the DA. The physical nature of the transmission lines is also a huge factor. If a transmission line becomes constrained in the RT that was not modeled in the DA, it will cause RT prices to be higher, especially if in a very constrained area (load pocket). Imports and exports between the various ISO’s cause constraints along the “seams” of the grids. Fuel (mainly natural gas) on the different pipelines also plays a significant role in prices.

As I said, there are many, many reasons that cause these differences, and while general trends can be modeled with statistics, knowing the fundamental operations of the grid will allow you to predict the outcomes much better. Since I worked at PJM for many years, I was able to learn where all of the generators are located, what prices they offer, their physical characteristics, and their affects on the grid… what the limits on the transmission lines are… what the natgas pipeline prices can indicate… as well as many other fundamental reasons for how the prices are being set. An understanding of the actual pricing algorithims also helps. You could be the greatest mathematician in the world, but unless you have an understanding of what is fundamentally driving the prices, you will not do well trading in electricity markets. It is not like other markets, in which prices are more influenced by financial traders. With electricity, the RT price is always set by the physical operation of the market, which can not be influenced by financial traders.

Hope this helps a little, and please let me know if I can explain more. I could write for hours about this. If you don’t mind me asking, where are you interning? Name of firm is not necessary, but how about location, size of firm, and type (hedge fund/energy prop shop/utility/etc…)? What is your background?

This was a fantastic read. I would love to hear your analysis on FTR's.

 

Haha, well I am glad that someone actual reads those reports, because I used to write it. On a side note, Up-To's were marginal in DA 87% of the time, with INC/DEC being 9%. These are not normal conditions, as the popularity of Up-To's has increased DRAMATICALLY in the past couple years due to the discovery of market participants discovering that balancing operating reserve charges can be avoided with them. There are current ongoing discussions to start charging Up-To's, which would return their popularity to previous levels.

In your example, you say what if someone DEC bids 10GW with 5GW of load. I'm not sure if this is what you were implying, but the 10GW would not all clear. The amount of MW that can possible clear is based on the actual physical limitations of the network. I don't know if you meant that if you bid in an excessive amount, the amount up to the limit will clear...

You also say that the RT price would suck. As I said in my post, a virtual bid can not effect the price at a node in the RT. Bidding in a high number of DEC MW that are marginal at a node, however, may cause the DA to be high, which may result in the spread between the DA and RT to increase, if the RT remains low, but it will have absolutely no impact on the price being set in the RT. The RT price will only be set by physical generators (or import/export transactions during emergency conditions). So, having high bids will run up the price in DA, which could greatly impact your return on the virtual.

Not sure what you mean by futures/swaps affecting prices, as they have no effect on the PJM network. I'm not too familiar with their transactions, as they were out of my jurisdiction to monitor.

 
thompsonpsu:
What exactly do you want to know?

Just the daily thought process that goes into trading this product. I'm fairly green on most things dealing with the electricity market so other then knowing a definition of what an FTR is, I'm not all that knowledgeable on what are the pros and cons to trading them. It seems there are some niche players in this space that focus exclusively on FTRs. Do FTR's behave somewhat like options? Are they actively traded? Is the trading more fundamental based vs quantitative?

 

FTR's, like virtual bidding, are used both by physical players as well as speculators. The physical guys (either severing load or supplying generation) use them to hedge their positions, while the speculators simply use them to profit without having any real assets.

An FTR is a contract to receive (or pay) the difference between the Day-Ahead Congestion Components (DA CONGCOMP) of two pricing nodes (Sink and Source), for a particular amount of MW cleared in an auction, for a particular length of time. FTR's are bought and sold in multiple rounds for each delivery period. The delivery periods are 24-hour, Peak, and Off-Peak... meaning that the 24-hour FTR is evaluated for all 24 hours of the day, while the Peak and Off-Peak are only evaluated on those hours (7am-11pm is Peak). You procure these FTR paths for a monthly or annual term, as well as Long-Term auctions, which gives a 3-year commitment. There is also a secondary market in which participants can trade FTR paths on a daily basis, but this is not an official auction.

A market participant can submit bids for any combination of two pnodes (load, gen, hub, aggregate, zone) on the grid, but must specify the direction (source and sink). For example, you can have your Source be a small 69kV substation in any region of PJM, and the Sink be the Western Hub. You could have a source be the PECO zone and the sink as COMED zone. You could also have a source at the PSEG zone, and the sink at any substation on the grid.

For any auction, you are submitting the maximum price you are willing to pay for that right of the path (hence, Financial Transmission RIGHT), in terms of total dollars. For example, you may bid $50,000 for 10 MW to have the right on the path of SubstationABC -> WESTERNHUB (Source->Sink). Bids clear so that the maximum amount of dollars is procured by PJM (the highest bid wins the contract). There can be multiple winners of the FTR up to the amount of MW allowed on each path. 5 different participants could own the right on SubstationABC -> WESTERNHUB.

Let’s use the previous scenario for the following example (Contract for SubstationABC -> WESTERNHUB @ 10MW for March 2013 Peak, and paid $50,000): The holder of the FTR path gets paid based on the (DACONGCOMPsink - DACONGCOMPsource)MW for each peak hour in the month of March. Say for a particular hour, the DA congestion component at WESTERNHUB is $100/MWh, and the DA congestion component at SubstationABC is $25/MWh. The holder of the FTR receives $75/MWh10MW = $750 for that hour. This is calculated for each hour in which the contract is subject to.

A participant’s gross revenue on the FTR is the sum of every hour across the period. Remember, however, that this participant paid $50,000 for the FTR. Therefore, the net profit is that total sum minus the amount paid for contract. So essentially this participant traded based on the assumption that the total revenue received from the congestion between the two points would be greater than the amount paid for the FTR. Conversely, if there is less congestion than expected during the time period, the participant would lose money.

There are also FTR’s that work in the opposite way, called Counterflow FTR’s. You are essentially selling a right on a path, and you get paid the money cleared in the auction. Using the same example, but from the selling standpoint, you would get paid $50,000 upfront for taking that obligation. You are then CHARGED for each hour of congestion. In this case, you are trading on the assumption that congestion on the path will be LESS than expected (the amount you will be charged from the hourly calculations is less than amount received from auction).

Physical players with actual deliverables (generation or load), use FTR’s to hedge their exposure. Take an example that a participant owns a 1,000 MW generator. Their energy market revenue is paid based on the DA LMP at the generator’s substation (GenSubstationXYZ) and the MW produced. (Note that the DA LMP is the sum of the (Energy + Congestion Component + Loss Component LMP’s). FTR’s are evaluated ONLY on the difference in Congestion Components of the LMP between the source and sink.) Let’s say for a typical hour there is no congestion on the system, and the generator gets paid $40/MWh1000MWh = $40,000 (energy component is $40, congestion component is $0, and loss component is $0). Suddenly there is a transmission outage near the generator, and the Congestion Component of the LMP at the generator is now -$20/MWh (the congestion component of the generator’s LMP is negative because they are causing congestion on the grid by injecting power). Their revenue now becomes (assuming the energy and loss component stays constant) ($40 + -$20 + $0)/MWh1000MWh = $20,000. In order to hedge against this scenario, they may have procured an FTR in which their Source is GenSubstationXYZ and the Sink is the Zone in which the generator resides, for 1,000 MW. If the FTR was active for that hour, the revenue they would receive from the FTR is equal to the (Zone CONGCOMP – GenSubstationXYZ CONGCOMP)1000, or, ($0 - -$20)1000 = $20,000. Now, FTR revenue + Generator revenue = $40,000. The $20,000 received from the FTR offsets the lost revenue from having a negative congestion component, and is equal to the money received absent congestion ($40,000).

For financial speculators, you are trading FTR’s based purely on the expected price differences between the pnodes. An easy strategy is to have a Sink pnode in a really constrained area of the grid, where congestion LMP’s are typically high, and have a Source at a more stable, non-volatile point (like WESTERNHUB). These constrained areas of the grid are where electricity is hard to flow into (low voltage lines with not many sources leading into it), or have expensive generators. These are called “load pockets”. Remember that these FTR’s will only be profitable if the amount received from the congestion is more than the price paid for it. Obviously, as you would imagine, these “easy money paths” are valued more to the market, and hence will have more people bidding on them, and hence drive up the price paid in the auction. If everyone knows there is a huge arbitrage opportunity between the historically lowest priced pnode and the highest priced pnode, everyone will be bidding to win that contract.

Lastly, these contracts are eventually paid out by the money collected from real-time congestion payments. I won’t go into a lot of detail about this, but since the congestion that is expected by PJM when they model the market is not always equal to the amount of congestion that actually occurs, there are sometimes “FTR Revenue Adequacy” issues. This basically means that PJM did not collect enough money to pay all of the participants that have FTR’s. You may only receive 95% of the revenue you should have made. You can follow more here: http://www.pjm.com/committees-and-groups/issue-tracking/issue-tracking-…{759ECF2E-D3AE-4AC3-88FB-8D3E45CBA614}

There are a ton of other details in regards to how FTR’s work, but is not really needed to explain here. This is just a brief primer. Please let me know if any of that made sense, and if you want me to expand on anything. You can also contact me on LinkedIn: http://www.linkedin.com/in/energyanalyst

 
ACD_Trader:
Thanks Thompson! That was a great explanation. I'll have to re-read it a few times but you explain things very concisely.

I knew they had optionality to them which is what caught my interest.

Thanks... I just wish I actually traded FTR's and Virtuals...

Further, when you refer to the options... "There are two FTR hedge type products: obligations and options. An obligation provides a credit, positive or negative, equal to the product of the FTR MW and the congestion price difference between FTR sink (destination) and source (origin) that occurs in the Day-Ahead Energy Market. An option provides only positive credits and options are available for only a subset of the possible FTR transmission paths."

-From PJM State of the Market Report: http://www.monitoringanalytics.com/reports/PJM_State_of_the_Market/2010… (this is where I used to work and write these reports):

 

Well...I still "sort of" disagree about virtuals not affecting RT prices. My point was this. If you have a gas generator running because the price curve bids in the DA were high enough to commit it for the entire onpk, there is a lot more reg-up available to the market as long as the unit stays running. If prices are lower, but not significantly lower, the unit will likely remain on minimums, because the operator won't want to turn it off unless instructed to do so by PJM. SO, say someone decides to export 1gw at the peak of demand. There are much higher chances that the price won't pop into oil (165+) because that gas gen can ramp quickly to meet the quick increase in demand. Overall, higher DA decreases volatility in the real time.

SO, while very true that RT prices will always be based on the physical marginal units, the mix of those units can be greatly affected by the DA market, which can be altered via virtual activity.

Your second comment about not really monitoring the futures/swaps market -- This is something that I personally wish PJM would monitor much more, because there are a lot of "head scratchers" that happen out there during volatile times. The PJM futures market is by far the most liquid/traded electricity product, and it is my belief that futures positions very much so affect market participant activity.

As far as FTRs, they're an interesting product, but a lot of speculators are backing out of the actual product with PJM and moving to OTC products. I'm not sure how much the OP knows about FTRs, but recently PJM has been "short-paying" FTR holders, citing that they haven't modeled external flowgates and loop flows very effectively during their auction process. What this means is this -- PJM isn't taking in enough money to cover their obligations per the awarded FTRs, so they basically say "oh well, we're not paying." As a speculator, you can't hedge for this, so you just don't trade it. What I've seen recently is OTC look-alike products, where counterparties trade at the PJM FTR clear prices, but it is a bilateral trade. That's where I see that market headed currently, unless PJM can get their act together.

 
 

Oversold, I understand your example, but I think you are assuming that virtual activity can influence actual unit commitment. This is not true. When PJM commits units in the DA, they do not look at non-physical trades. The overall market will clear with the virtual activity, but PJM will also run a separate case in order to actually schedule units. So say someone DECs a bunch, which causes a small CT to run... this unit may not actually be committed to run in RT (but would look like it when looking at DA schedules). When I worked at PJM, one thing I was involved with was examining the dispatch optimization methods and unit commitments. I had access to the actual programs being used to solve the markets, and had to compare actual RT dispatch decisions to DA optimization ("Perfect Dispatch", if you are familiar). So, unless I'm remembering things incorrectly, I still disagree with virtuals directly influencing RT prices.

Regarding your comment on futures/swaps... you're talking about NYMEX/ICE, correct? If so, PJM is not allowed to monitor these because it is out of their jurisdiction. Since they are not traded directly through PJM, they do not have access to that data. The products are subject to The FERC and the CFTC. I'm not sure what the final ruling was, but when I was at PJM, there were motions to have the CFTC monitor FTR's. These discussions were part of Dodd-Frank. I think the main argument was whether they were OTC products, and what organization has jurisdiction. When I was there, we definitely petitioned to The FERC, NYMEX, CFTC, ICE to grant us access to monitor this activity, but nothing ever happened. The biggest flaw with this is that physical players can DEFINITELY influence the PJM prices which could cause their NYMEX/ICE products to profit... and no one would ever know. Say a large generating company has a ton of futures on NYMEX that are indexed to a zonal price. The company obviously knows when their units are going to be on outages, which will drive up the prices. They could have a ton of contracts on NYMEX based on this knowledge. Unfortunately, PJM only has access to data for the activity within PJM, so there is no way to see if a company is intentionally influencing prices to drive up OTC products. And conversely, people at the CFTC/NYMEX/SEC or whomever don't have access to PJM data. Additionally, anyone there has no idea how to interpret the PJM data. It's a complicated issue.

Regarding the FTR comment. Agree that FTR is a shaky product due to revenue inadequacy. It has gotten much worse over the years. The worst I've seen it is when the MT STORM - PRUNTYTOWN (I believe) went out of service halfway through the year. However, I disagree with your notion that it is entirely PJM's fault. They do not have a crystal ball, and can not fully predict the future of the transmission lines. Like I said, a transmission line could suddenly go out midway through a year, which can't possibly be modeled when the market clears. I would more so blame the load serving entities, since THEY own the transmission lines, and would have a better idea on their future. They also have better knowledge regarding their predicted load, line ratings, outages, etc... So it's really up to them to provide better information so PJM can accurately clear the market.

Good discussions on this forum. Keep it up.

 

Hi, Thompson. Your introductions are really helpful!! Thanks a lot. Do you happen to know some facts of generators, utilities on how they hedge against risk. Do they involve in FTRs more in their portfolio or do they also involve in virtual trade as a way to ensure their profits? Thanks

 

To further clarify on my example of virtual activity not affecting actual unit commitment. Think about this: say a participant increment offers 1000 MW at the generator substation of a cheap baseload power plant and clears, since the offer is less than the offer from the generator. Do you really think that the coal plant will not run in the Real Time? Unlikely. This would cause all sorts of problems with grid reliability, and hence is why virtuals will not affect the actual scheduling of units.

 

Hey thompsonpsu, would like to thank you for the very informative threads about incremental bids and financial transmission rights; your posts tremendously increased my knowledge on power markets.

If you have the time, could you explain the concepts behind up-to-congestion and the processes behind them for market and grid operators ?

I am currently reading "electricity markets: pricing,structures and economics" by Chris Harris to learn more about power markets but your posts are somehow more practical and easy to understand.

Thanks again your your previous posts and looking forward for the next ones !

 

thompsonpsu, thank you so much for your extremely detailed responses. I apologize for not responding earlier, I don't get e-mail notifications when people respond.

You've restored my faith in humanity thompsonpsu, it's amazing that you're willing to give such a detailed response to a stranger on the Internet. Will be in contact with you in the near future.

 

If you join an IPP as an entry level real time power trader, how much responsibility will you have to start? Meaning, how large will your book size be, i.e. how many MW's will you most likely be trading accountable for managing in a 12 hour shift?

 

I am in this ISO business for years, and worked for a well-known software vendor to help many participants in different ISOs. I must say, your fundamentals are absolutely clear on these topics. I loved reading your posts. Thanks all to be part of this discussion. It’s very informative.

 

I believe so. A genenator can either choose to sell power directly or to "standby" the capacity to provide grid reliability in case of supply shortage. In the regions where a capacity market exits, capacity can be bought through auction and traded.

 

I trade power in Texas (ERCOT) and I can tell you that the term 'Capacity' here refers to Ancillary Services. There are two primary markets: the Day Ahead Market and the Real Time Market. In the Day Ahead Market, the state estimator determines how much total energy will be needed for the state's consumption and assigns each market entity a required amount of capacity mws (according to the firm's geneerator output capacity sum). Since supply will never equal demand, and since all generation units have differing ramping capabilities, Responsive Reserve and Non-Spin Capacity is required in case the delta between load and generation deviate tremendously due to an Energy Emergency and is deployed to prevent generator and load trippage. Regulation Up and Regulation Down are both capacity energy used to smooth out ramp ramps as well as control small deviations between gnerator and load that are all part of normal operations.

In terms of a car: Think of Nonspin and Responsive Reseve as the power capacity your engine has but is not in use and may only be used in emergencies. Think of Regulation up and down as pumping the gas to make sure you maintain a somewhat steady speed.

 

There’s an energy cost exposure so Retail Electric Providers (REPs) must acquire energy to serve load. You have bilateral trades, DAM trades and real time trades mainly ICE trades. The more forward ahead the energy is purchased the higher the price. Not all deals are capacity trades ERCOT is an energy only market so DA virtual deals are considered fixed financial deals. Some deals are physical (ancillary service) meaning the energy has to be delivered by the counterparty or asset and other are a financial buy. Some are Energy. In terms of P&L you take the DART spread. Day ahead prices vs. real time prices. The trader will benefit from a long position if prices appreciate and the trader will benefit from a short position if the prices depreciate. You have your Point to Points (PTPs) where you profit from the basis of two nodes DA/RT. Point-to-Point Obligation - Can result in a payment or charge. Depending on source and sink - respectively origin/destination.

The purest form of giving is anonymous to anonymous..
 

Dicta voluptatem quis dolores aliquam. Ipsa deleniti aut quam.

Provident alias doloremque laudantium tempora voluptatum cupiditate. Eos atque quia reprehenderit est. Commodi quibusdam sunt ducimus perspiciatis. Vel eaque cum error. Magnam ea et doloribus praesentium.

Recusandae sunt illum et et voluptatibus nisi repellat. Consequatur nihil omnis praesentium accusantium facilis aut sed. Magnam molestias laudantium impedit.

 

Nihil molestiae assumenda quas et rerum corporis. Fuga dicta voluptatem dolore commodi voluptatem cupiditate. Fugiat accusantium non minima aliquid.

Consequatur blanditiis id doloribus ut omnis quisquam. Repudiandae voluptatibus tenetur et. Aut autem reiciendis rerum voluptatem repellendus exercitationem est. Accusamus dolores illo quod et inventore est ut.

Quibusdam nobis et velit aut dolores porro. Ad illum repellat nobis ducimus ipsa. Vel sit a commodi provident porro earum.

Nulla consequuntur eum qui saepe qui omnis magni. Et sequi aut saepe aliquid tenetur voluptatum eum. Facilis itaque porro qui magni ad. Eius at autem vel aut possimus quia reiciendis voluptatem. Praesentium id nihil id rerum. Exercitationem aperiam dolorum exercitationem quisquam assumenda voluptatibus doloribus est.

Career Advancement Opportunities

April 2024 Investment Banking

  • Jefferies & Company 02 99.4%
  • Goldman Sachs 19 98.8%
  • Harris Williams & Co. New 98.3%
  • Lazard Freres 02 97.7%
  • JPMorgan Chase 03 97.1%

Overall Employee Satisfaction

April 2024 Investment Banking

  • Harris Williams & Co. 18 99.4%
  • JPMorgan Chase 10 98.8%
  • Lazard Freres 05 98.3%
  • Morgan Stanley 07 97.7%
  • William Blair 03 97.1%

Professional Growth Opportunities

April 2024 Investment Banking

  • Lazard Freres 01 99.4%
  • Jefferies & Company 02 98.8%
  • Goldman Sachs 17 98.3%
  • Moelis & Company 07 97.7%
  • JPMorgan Chase 05 97.1%

Total Avg Compensation

April 2024 Investment Banking

  • Director/MD (5) $648
  • Vice President (19) $385
  • Associates (86) $261
  • 3rd+ Year Analyst (14) $181
  • Intern/Summer Associate (33) $170
  • 2nd Year Analyst (66) $168
  • 1st Year Analyst (205) $159
  • Intern/Summer Analyst (145) $101
notes
16 IB Interviews Notes

“... there’s no excuse to not take advantage of the resources out there available to you. Best value for your $ are the...”

Leaderboard

1
redever's picture
redever
99.2
2
Betsy Massar's picture
Betsy Massar
99.0
3
BankonBanking's picture
BankonBanking
99.0
4
Secyh62's picture
Secyh62
99.0
5
dosk17's picture
dosk17
98.9
6
GameTheory's picture
GameTheory
98.9
7
CompBanker's picture
CompBanker
98.9
8
kanon's picture
kanon
98.9
9
bolo up's picture
bolo up
98.8
10
Jamoldo's picture
Jamoldo
98.8
success
From 10 rejections to 1 dream investment banking internship

“... I believe it was the single biggest reason why I ended up with an offer...”