Competition in the Electric Industry: Emerging Issues, Opportunities, and Risks for Facility Operators (1996)

Jim Clarkson

Heath Petra Resources Company

Full-fledged competition and the ability to buy on the wholesale market may be a few years away, but large power purchasers already have more options than they had in the past. Utilities, as competition approaches, are eager to make deals and are trying to become more responsive to their big users. From the customers' standpoint, now is the time to press the advantage.

About 44 percent of the power generated in the United States goes through the bulk wholesale markets before reaching the retail customer, so there is an active, well-established trading mechanism already. Any problems that might have existed with utilities' swapping power have already been worked out. The commercial arrangements are also well developed.

The prices at which these trades are made are sometimes startlingly low, as low as $3 per megawatt hour (0.3 cents per kilowatt hour). Yet the generators make the sales because they contribute to the variable costs of running the company. Similarly, a manufacturing company producing widgets at a break even price of $1.00 will keep selling widgets at a price of, say, $0.60 if it can make the units for $0.50 in variable cost. In both cases the costs are divided between fixed costs, which are amortized over time, and variable costs to make the unit. The sales cover the variable costs and contribute to the recovery of fixed costs.

This phenomenon is known throughout the rest of the economy, but not to the utility industry. In the competitive sector of the power industry, prices are down near the variable costs of the utilities. Those prices represent short-term bargains, which would be akin to the widget manufacturer trying to eliminate its inventory and selling widgets for $0.25. In the West particularly, unusually high rainfall in 1995 filled up the hydroelectric reservoirs, and power is being given away at very low prices. In addition, some nuclear plants can make electricity at costs below 1 cent per kilowatt-hour, and the owners have been unable to sell all of that power until recently. Prices are below what one would expect.

In an economy with overcapacity, one would expect that, as the need for power increases, the generation resources with the next highest variable cost will be called on at each step. If the amount of power needed required all of the generating plants available—all the way up to the ones costing 18 mils or 1.8 cents per kilowatt-hour to make power—then that would be the price for everyone, even for hydroelectric plants, which cost virtually nothing to run.

That is the way our market works. It is a new experience for the utilities, who have never experienced periods in which they were not collecting on their fixed costs. Their fixed costs are having to be deferred unless they can get their variable costs below the marginal cost of bringing on the next unit.

Retail electricity prices today vary widely from utility to utility in this country, between 3 and 5 cents per kilowatt-hour, and more in some places. The fully loaded costs in the competitive wholesale market are much less than that. Resources in the range of 0.3 to 0.9 cents per kilowatt hour are not going to be available for long, but in the future, supply and demand will stabilize at a price of about 1.5 to 2 cents.

The retail market, when it is in place, will go to those same prices, and there will be less variation, because fixed costs (which led to much of the variation) will not come into play in setting prices. The utilities, in selling power, need to receive, say, 2 cents to cover their variable costs. They are also collecting capital costs in the form of demand charges of, say, $10 or $12 per month. But they will forego recovery of capital costs in a competitive market, so the price of power will be about the same to

customers with low load factors, such as residential customers, as it is for industrial customers. The large difference in prices to the two classes of customers today is due to the way fixed costs are allocated over the units sold. A big user using many units on a steady basis has the fixed costs spread over more units, so the average cost is perhaps 3 to 4 cents, while a residential customer, with a worse load factor, might be charged 7 or 8 cents. In a competitive market no one collects the fixed costs, and the variable costs apply to everyone, so price variation is smaller.

This low-cost power will create a boom in certain industries. Some of America's electricity-intensive industry will come back to the United States, at least into the areas that first implement competition.

The aluminum industry is an example. For 20 years the conventional wisdom held that one cannot build an aluminum plant in the United States, because of the high electricity costs. About a year ago I was involved in assessing a U.S. aluminum company's plans to build a smelter in Malaysia. An aluminum smelter needs a huge amount of power, so a power plant was going to have to be built, in this case a plant fired by low-grade coal. The capital cost was estimated at $800 per kilowatt, and coal was expected to cost about $15 to $17 per ton. In all, it came to about 28 mils, or 2.8 cents, per kilowatt-hour for power. Now the company appears to have abandoned that plan, and is considering expanding in the United States, because of the prospect of power selling for 1.5 to 2 cents per kilowatt-hour. The United States, in other words, has a competitive advantage in electricity prices. Elsewhere in the world, if one intends to build a facility with a large power load, one must build one's own power plant. The United States has about 20 percent excess power, and power needs here are growing at about 2 percent per year. As a result, we can expect fire-sale prices for electricity for a number of years.

There is more good news. In a few years, when excess power is used up and the utilities can begin raising prices, due to developing shortages, they will find that technology has placed a ceiling on prices. That ceiling is due to the technology of combined-cycle generation (a natural-gas fired turbine, with waste heat recovered to power a steam turbine). The cost of these highly efficient systems has fallen dramatically in the past few years, and continues to drop. One can expect electricity from combined-

cycle plants to be priced at 2.4 to 2.8 cents per kilowatt-hour. With the marginal cost of electricity at that level, future prices should not rise to the levels of regulated retail prices in the past.

Meanwhile customers can use interruptible power to keep power costs lower. That is, they can agree to interrupt their power at signals from their utilities (for example, at times of peak load) in return for lower rates. Customers can cut off some of their equipment or use local generators to make up for the interrupted power. In effect, the utilities in these arrangements are giving customers power at its energy cost, without collecting demand charges.

Many utilities also offer customers so-called real-time pricing. In these arrangements, the price rises and falls to reflect the actual cost of power at the moment. Customers receive strong cost signals, and can schedule their energy use to take advantage of low-cost resources (such as in off-peak hours). For the first time there is feedback between suppliers and users. Users have an opportunity to control their own costs. Utilities can communicate in the strongest possible terms the cost of starting up that last peak load power plant, and can even influence customers consumption enough to avoid adding new peaking capacity altogether.

In practice, however, many utilities have used real-time pricing as a means of adding to their revenue without adding plants. The price signals in such cases are artificially high, reflecting not the cost of power on the grid, but the utilities' desire for revenue. Niagara Mohawk and a few others have developed revenue-neutral real-time pricing tariffs.

Still, if a utility offers real-time pricing, customers should examine the option; these time-of-day premiums are negotiable as to exactly the sort of premium that is to be added to the overall price. Furthermore, even if the price spikes are artificially high, customers can work around them enough to lower their overall power costs. Such activity can result in enormously profitable paybacks.