AVOID VOLATILITY, BUT PAY A LITTLE MORE.

Under deregulation, some retail power customers are applying power purchasing techniques commonly used by utilities to avoid feeling wholesale price volatility at the retail level. One such technique involves “laddering” contracts. To ensure acceptable results, a clear understanding is needed of this technique’s limitations.

HOW THE BIG BOYS DO IT

Utilities have a reasonably predictable load and purchase power from multiple suppliers in blocks that together meet that load. Such blocks cover different terms (e.g., time periods) and volumes (e.g., kWh). When laid out as a chart, the borders of these blocks look to some like ladders.

By varying term lengths, any significant price jump for part of a load during one period is mitigated by the price certainty of fixed price blocks (for a portion of the same load) whose terms overlap with that of the higher-priced block. The value of any price drop is, however, muted by that same fixed price block. The net result is smoother pricing: not very high, not very low, despite market conditions.

APPLYING AT THE RETAIL LEVEL

The laddering process as used by utilities works when power is purchased and allocated to a large general load and balanced by utility generation, but it may not work as well for a few retail power accounts. Power use must be balanced so that supply and demand are equal at all times: unless facilities have on-site generation that automatically adjusts to maintain a consistent demand, end users’ power needs are constantly varying.

To deal with such variations, utilities - or independent system operators (ISO) that operate wholesale grids - act as balancing agents. If more power is used than covered by a contract, the excess is provided at cost by the utility or ISO. Multiple retail suppliers trying to balance usage variations in concert with multiple contracts, all for one account, is not practical. As a result, deregulated retail markets generally require that only one supplier provide power to a single account at any one time.

TWO WAYS TO PLAY THE GAME

There are at least two ways to adapt this technique for retail use. The first involves serving a portfolio of accounts from multiple suppliers, while the other involves serving a single large account with multiple contracts from one supplier. Suppose, for example, we have four suppliers each fully serving specific accounts that, in four groups, each make up 25% of the total load portfolio. Each supplier’s contract has a different term length. At startup, this method requires buying 75% of the power at prices higher than the lowest of the four bidders. As bids vary in the future, however, overlapping the contract terms results in lower price volatility across the portfolio, but not necessarily for a specific account. The ability to secure the lowest price and the risk of getting stuck with a higher price have both been avoided to secure price consistency for the portfolio.

Under the second scenario, the load for one large account is served by four separate contracts (sometimes called tranches) all with one supplier. Each has a different term and a different price. At the start of this process, each contract covers 25% of the load. In the chart, terms for the contracts are: A – six months, B – 12 months, C – 18 months, and D – 24 months. New bids (always from the same supplier) must be secured every six months for one of the four tranches. After the first six months, contract A is rebid (to the same supplier) to become contract A1. While it is priced at immediate market conditions, price for the remaining 75% of the load is secured by contracts B, C, and D. Even if A1’s price is much higher, it affects only 25% of the account’s load, thereby minimizing impact on the account’s total cost. As seen in the chart, retail pricing varied from $.035 to $.080 (230% variation), but the average price paid was between $.05625 and $.06875 (22% variation). Lower volatility was secured, but the lowest price may have been missed because no price competition occurred.

Under either scenario, the customer must re-bid part of his load several times a year. In the second scenario, contractual benchmarks may be needed so that the single supplier cannot ratchet up his pricing knowing that the customer is tied to him.