About 30 states have begun (em

either through the legislature, the utility commission, informal working groups, or some combination of these (em to consider issues such as retail wheeling,...

gas and oil may be burned in units with very different characteristics and corresponding costs. For gas-fired units, the two primary unit types are combustion turbine (CT) units and combined-cycle (CC) units. CTs are peaking units that are relatively inefficient and CCs are more efficient midmerit units. For oil-fired units, the two primary unit types are CTs and steam units. The efficiency of fuel-burning generating units is generally measured by the heat rate, in BTU of fuel input per kilowatt-hour of output. A higher heat rate means that a unit is less efficient and a lower heat rate means that a unit is more efficient. As a result of the differences in fuel efficiency across unit types, the assumptions made about the marginal unit type in the Spinner article have a significant impact on the conclusions of the paper. The incorrect conclusions drawn in the Spinner article stem largely from these faulty assumptions.

The Spinner article incorrectly assumes that when gas is the marginal fuel, more efficient gas-fired combined-cycle units are on the margin most of the time. The article assumes that less efficient gas-fired combustion turbine units are on the margin only in the months of January, December, June, July, and August, and only when hourly load is greater than or equal to 95 percent of the maximum monthly load for the two winter months and 90 percent for the three summer months. Spinner’s assumption that more efficient CCs are generally on the margin when gas is the marginal fuel is wrong. In fact, gas-fired CTs were on the margin in 25,499 five-minute intervals in 2005 (24 percent of all intervals) while the Spinner article assumes that gas-fired CTs were on the margin in only 1,357 intervals (1 percent of all intervals). This information was published in the 2005 State of the Market Report and was also communicated to Spinner in April 2006.

Assuming that CCs are on the margin when gas is the marginal fuel is equivalent to assuming that costs are lower than they are. Similarly, assuming that oil-fired steam units are on the margin when oil is the marginal fuel is equivalent to assuming that costs are lower than they are. Spinner uses a heat rate of 7,500 Btu/kWh for gas-fired CC units and 12,000 Btu/kWh for gas-fired CT units. These assumptions are significant because, using these heat rates, the fuel component of a CT’s marginal cost is 60 percent higher than that of a CC. In addition, the Spinner article incorrectly assumes that when oil is the marginal fuel, an oil-fired unit with a heat rate of 11,000 is on the margin. In fact, the heat rate for oil-fired steam units is about 12,600 and for oil-fired CTs is about 14,100. Spinner’s mistaken assumptions about the characteristics of units on the margin significantly impact his conclusions by understating marginal costs.

Spinner erroneously suggests that his assumptions are less significant because the observed relationship between estimated costs and prices is stable for 18 months and changes only during the latter half of 2005. In fact, the apparent difference