RTP assumes that price spikes will deter load. But how will customers behave if they've hedged against that risk?
Tomorrow's electricity industry promises a wealth of pricing options as wholesale generation becomes more like a commodity. Spot pricing marks one example. And with spot markets will come a greater need for price derivatives (em hedge contracts that will permit customers to trade or shed risk to achieve a higher degree of price certainty.
This assumption poses a critical issue: How will price derivatives affect the behavior of customers who already subscribe to real-time pricing? Simple intuition would tell us that customers would turn to derivatives and hedge contracts to avoid having to curtail consumption to save costs during those periods when real-time prices spike upwards in response to market conditions. In other words, real-time pricing might fail to deter load at critical times. If that assumption should prove correct, utilities would need to take account of such behavior in their generation and transmission system planning.
Nevertheless, things aren't always what they seem.
Based on research that I have conducted at Georgia Power Co., I assert that, contrary to intuition, RTP customers will not change their usage habits simply because they have purchased a price-risk derivative contract. The analysis that follows will attempt to explain the economic logic for this seemingly counter-intuitive result.
The key lies in the fact that price derivatives focus on the average RTP price over the time period and not on specific interval prices within the time period. (This phenomenon is similar to what's referred to as an "Asian Option" in commodity futures trading.) The logic can prove daunting, however.
Simply put, derivatives don't change the economic value of the product of the energy input (em they merely add to or diminish the resulting economic effect of producing or not producing. And, the determining factor on whether to produce is presumably based upon the cost input represented by the spot price of electricity, all else equal. In other words, if it were economic to produce without a derivative, it is even more so with a derivative. Similarly, if
economics supported a curtailment absent any derivative contract, it should be even more cost-effective to curtail with a derivative in place, since the value of the product is controlled by exogenous factors apart from the input price of electricity.
Others may differ. They may believe that customers will not understand these economic comparisons and will react "from the gut," based upon the more intuitive notion of purchasing derivatives to avoid having to curtail consumption during RTP spikes. I acknowledge that at their first encounter with high prices (em for a brief period, at least (em customers may possibly react by not curtailing consumption. However, my experience with RTP indicates that customers eventually will reach the same economic conclusions that we have at Georgia Power. Once that happens customers will change their behavior at breakneck speed, accommodating the use of financial derivatives to the real benefits still available under real-time pricing.
Product Choice:
Weighing Value Against Price
Students of college economics will recognize the standard demand curve, with its downward slope from upper left to lower right, illustrating changes in consumption as a function of price. In this case, to measure customer response to a range of changing real-time prices, the analysis assumes a "kinked" demand curve. The kink represents that section of the demand curve that appears vertical, indicating that consumption remains constant over a given range of real-time prices.
This idea of a kinked demand curve underlies the base-case scenario, which illustrates how a hypothetical customer might migrate between three different electric supply products, as real-time prices rise and fall above or below the economic value that the customer assigns to those products.
For instance, assume that a customer can choose among three electric supply products, each exhibiting a different level of net economic value for the customer. The customer will remain indifferent as the real-time electricity price rises up to this level, but will curtail the product when price exceeds value. (See figure 1. Note also: While the examples will assume that an all-or-nothing curtailment brings the customer's consumption immediately to zero, in practice an actual curtailment of product would require a gradual reduction in load supporting the product.)
Hedging Risk:
Buying a CAP on Real-Time Prices
Assume the customer continues to accept delivery of Product Line B, at an expected real-time price of 7¢/kWh, with its assumed inherent economic value of 15¢/kWh. Then assume further that for an additional cost of 1¢/kWh, bringing the expected product cost up to 8¢ %n1%n, the customer is offered the option of buying a derivative (em a price cap at 13¢/kWh. Note that this price cap (and any other derivative, for that matter) will apply to the average real-time price in force over a given time period. Individual RTP prices can, therefore, exceed the CAP without the CAP being initiated; only if the average RTP price over the period exceeds the CAP will it be initiated.
This example can be illustrated (see Figure 2) in a graph that shows the effective cost of the price cap, and both the expected and guaranteed minimum economic values for the product after the customer has purchased the derivative.
Consumer Behavior:
How Derivatives Alter Load
A quick examination of the test case scenario by way of a decision matrix (see Table 1) leads directly to some propositions and a conclusion about how the purchase of derivatives will affect consumer behavior and alter load.
Proposition A. Assuming the real-time price remains below 15¢/kWh (the economic value of Product Line B), the customer will run Product Line B and receive at least the economic value represented by Area III, less Area IV. (Note: Area IV is a sunk cost after CAP is purchased and will not affect run decision.)
Proposition B. If the RTP price exceeds 15¢, the customer will curtail Product Line B and avoid a loss of product value. (Without curtailment, the customer would lose value equal to RTP-15¢ %n2%n.) Moreover, the decision to curtail will also reap a benefit for the customer (em the receipt of compensation under a settlement of the derivative contract, equal to
(RTP -P) - (C1-P2) %n3%n per kWh.
Conclusion: Even when the customer has purchased a price cap, the customer will continue to curtail consumption as if no derivatives existed.
Now suppose that instead of being offered a price cap of P3 (13¢/kWh), the customer had requested a price CAP of 15¢/kWh. In that case, the analysis leads to two additional propositions:
Proposition C. The customer will run Product Line B as long as the price falls below 15¢/kWh (the economic value of the product). This behavior is identical to the decision the customer would reach without a price cap.
Proposition D. If the RTP price exceeds 15¢/kWh, the customer will curtail Product B just as he would have done without the price cap. But with the CAP, the customer is assured that by curtailing consumption, his economic value for each individual curtailment decision will always be greater with the CAP than without the CAP.
Example:
• Customer needs 300 kWh in hour 1. It takes 100 kWh each to produce Product A, Product B and Product C.
• Economic values of Product A, B and C are 5¢, 15¢, and 50¢ per kWh, respectively.
• The customer is offered a price cap that will cost a premium of $1 for 100 kWh.
Table 2 illustrates this example. It compares customer behavior with and without price caps of 10¢/kWh and 15¢/kWh, across a range of prices. Whether the price is high or low, and whether it exceeds or stays below the price cap, the customer will always maximize volume by curtailing consumption in response to price in the same manner with the price cap as without it.
Some Thoughts to Ponder
If a customer is not going to operate any differently with price derivatives, then why would he purchase them when the expected cost with them would exceed the expected cost without purchasing a derivative?
Several possible answers come to mind:
• To levelize costs over time.
• To minimize inopportune timing of high price events.
• To eliminate catastrophic, unusual possibilities.
• To lock in profitability.
This analysis leads one to draw the following conclusion:
Price derivatives do not change curtailment decisions for high-price events. However, they do lock in profitability, eliminating the possibility of profit erosion from a cost increase when the decision is made not to curtail service. t
Mike O'Sheasy is rate manager of the marketing department at Georgia Power Co.
Figure 1 (em Base Case Scenario (not included)
Figure 2 (em Test Case Scenario (not included)
Figure 3 (em Multiple RTP Price Scenarios (not included)
Table 1. Pricing Scenarios
With and Without a Price Cap
Low Price High Price Uneconomic Price
RTP < P3 P1 > RTP > P3 RTP > P1
(Actual price stays under (Actual price exceeds price (Actual price exceeds value the price cap.) cap, but remains below of Product B. Also exceeds value of Product B.) price cap.)
With Hedge Behavior: Customer runs Behavior: Customer runs Behavior: Customer curtails
CAP = P3 Product Line B Product Line B Product Line B
(Customer response Cost: Pays real-time price Cost: Pays capped price Cost: Pays cost of CAP (1¢) after buying plus cost of CAP (1¢) (13¢) plus cost of Benefit: Avoids loss of product derivative.) CAP (1¢) value (extent that RTP exceeds 15¢); plus receives settlement on hedge (RTP minus 13¢-CAP)
No Hedge Behavior: Customer runs Behavior: Customer runs Behavior: Customer curtails
CAP = n/a Product Line B Product Line B Product Line B
(Customer response Cost: Pays real-time price Cost: Pays real-time price Cost: Zero without buying Benefit: Avoids loss of product derivative.) value (extent that RTP exceeds 15¢)
Table 2. Pricing Scenarios
Comparing Different Price Caps
RTP Price = 8¢ RTP Price = 12¢ RTP Price = 20¢
With Hedge Behavior: Customer runs Behavior: Customer runs Behavior: Customer curtails
(Customer response Product Line B Product Line B Product Line B
after buying price Cost: Pays 8¢/kWh plus a Cost: Pays 10¢/kWh plus Benefit: Avoids loss of product
cap of 10¢/kWh.) $1 up-front premium $1 premium value of (20-15¢/kWh) x 100 kWh
for CAP Benefit: Derived value = = $5, plus enjoys CAP compen- Benefit: Derived value = 100 kWh x (.15 - .10) - sation of (20-10¢/kWh) x
100 kWh x (.15 - .08) - $1 = $4 100 kWh - $1.00 %n4%n = $9 %n5%n
$1 = $6
No Hedge Behavior: Customer runs Behavior: Customer runs Behavior: Customer curtails
(Customer response Product Line B Product Line B Product Line B
without buying Cost: Pays 8¢/kWh Cost: Pays 12¢/kWh Benefit: Avoids loss of product
price cap.) Benefit: Derived value = Benefit: Derived value = value of (20-15¢/kWh) x 100 kWh
100 kWh x (.15 - .08) = $7 100 kWh x (.15 - .12) = $3 = $5
With Hedge Behavior: Customer runs Product Line B Behavior: Customer curtails
(Customer response Cost: Pays real-time price plus premium of $1 Product Line B
after buying price Benefit: $6 and $4, respectively (see above) Benefit: Avoids loss of product
cap of 15¢/kWh.) value of (20-15¢/kWh) x 100 Kwh
= $5, plus enjoys CAP compen- sation of 100 x (20¢ - 15¢) - $1.00
(premium) = $4
No Hedge
(Customer response Behavior: Customer runs Product Line B Behavior: Customer curtails
without buying price Cost: Pays real-time price Product Line B
cap.) Benefit: $7 and $3, respectively (see above) Cost: Zero
Benefit: Avoids loss of product
value (extent that RTP exceeds 15¢)
1The assumption is made here that, after paying a premium for the price-cap, the customer would not run up against any internal budget limit for purchasing electricity.
2The current curtailment value for Product Line B.
3Derivative value resulting from the purchase of the price cap.
4The premium cost of $1.00 for 100 kWhs for the CAP represents a sunk cost after it is incurred, and should never thereafter affect the decision of whether to run or curtail Production Line B. An implicit assumption is made here that the customer doesn't begin with a fixed energy budget that would now be reduced after paying the CAP premium, such that any future incremental purchasing decisions would be affected.
5Had the customer not curtailed consumption, its derived value would have been (15-20) x 100 + (20-10) x 100 -$1 = $4.
Articles found on this page are available to Internet subscribers only. For more information about obtaining a username and password, please call our Customer Service Department at 1-800-368-5001.