# Switching Rate Schedules — Notice, Minimum Stay and Eligibility

> The modeling says switch. The tariff says how, when, and for how long you are committed. Reading the second document after the first is how a good analysis becomes a bad year.

Section: Rate Schedules and Tariffs  
Author: Nil Masferrer Jiménez  
Published: 2026-08-18  
Reading time: 4 min

The analysis is done, the model reconciles, and one schedule is clearly cheaper. That is the easy half.

The other half is a set of provisions in the tariff about how you move, when it takes effect, how long you are committed, and what happens if the answer changes. Reading those after electing rather than before is a common and expensive sequence.

## The five provisions to extract

| Provision | The question it answers | Why it matters |
| --- | --- | --- |
| Eligibility | Do you qualify, on what criteria, measured over what period? | Qualifying today is not the same as qualifying on a twelve-month test |
| Notice | How much warning, and does the election take effect at the next cycle or a later one? | Determines whether you catch the season you were aiming at |
| Minimum stay | How long are you committed? | Usually the largest single risk in the decision |
| Return conditions | Can you go back, and on what terms? | Some schedules restrict return for a period |
| Determinant carry-over | Do ratchets, contract demands or historical peaks follow you? | A ratchet that survives the switch can erase the benefit |

*What to find before electing a new schedule*

The last row is the one most often missed. A tariff comparison built on the assumption of a clean start is wrong if the new schedule's ratchet clause looks back at peaks recorded under the old one, and the difference can be the whole of the projected saving in the first year.

## Eligibility is measured, not asserted

Eligibility criteria are usually stated as thresholds — a maximum or minimum demand, a voltage level of service, a customer classification, occasionally a load factor — and they are typically measured over a period rather than at a moment.

A schedule available to customers whose demand has not exceeded a given level *in any of the preceding twelve months* is not available to a site that exceeded it once in March. Confirm the measurement basis, not just the number.

There is a second-order consequence for anyone planning demand work. If a demand reduction project would bring the site below an eligibility threshold for a more favorable schedule, that is a benefit of the project, and it belongs in the business case alongside the direct saving: [building the business case for demand reduction](https://thedemandcharge.com/articles/demand-reduction-business-case).

## Mandatory reassignment

Not all movement is voluntary. Where a schedule has a mandatory applicability clause — for example, service above a stated demand level *shall* be taken on a particular schedule — crossing the threshold triggers reassignment whether or not the customer asks.

This matters for planning. A site whose demand is growing toward a mandatory threshold should model the destination schedule **before** it arrives there, because the arrival will not be optional and the structure on the other side may be very different. Discovering it after the fact is how a growth project produces an unexplained step change in the electricity budget.

## Minimum stay is the real commitment

A minimum stay obligation exists to stop customers arbitraging between schedules — electing a demand-heavy schedule in winter and an energy-heavy one in summer, taking the best of both and paying for neither. It is a legitimate provision and it is also the largest risk in the decision.

Twelve months is common. During that period, if the load changes in a way that makes the new schedule unfavorable, you serve it out.

The defense is procedural: **model the alternative schedules against a changed load, not only against last year's.** Specifically, against the profile that would result from a planned expansion, a shutdown, a shift pattern change, or a demand reduction project already in the pipeline. If a schedule wins on the historical profile and loses on the planned one, the switch may be right and the timing wrong.

## Sequencing against a project

There is a genuine ordering problem when a demand reduction project and a tariff switch are both on the table, and there is no universal answer.

**Switch first** if the current schedule is badly wrong for the current load and the project is a year away. There is no reason to keep paying the wrong price while waiting.

**Project first** if the project will substantially change the load factor, and switching twice inside the minimum stay would be impossible. Change the load, observe the new profile for a few months, then elect on evidence.

**Model them together** in every other case. The right question is not which schedule suits today's load or tomorrow's, but which suits the load you will have for most of the minimum stay period.

## The election itself

**Electing a new schedule**

1. Confirm eligibility against the measurement basis in the tariff, using twelve months of history rather than the current month.
2. Confirm the notice period and the effective date, and count backward from the season you are trying to catch.
3. Confirm the minimum stay and the return conditions in writing.
4. Ask, in writing, how ratchets, contract demands and historical peaks are treated on transfer, and keep the answer.
5. Confirm which riders attach to the new schedule. They differ, and a lower base rate with a heavier rider set can be more expensive: [riders and surcharges](https://thedemandcharge.com/articles/riders-surcharges-explained).
6. Submit the election in writing, whatever the utility says is acceptable, and keep the acknowledgement.
7. Reconcile the first bill on the new schedule against your model, line by line. If they disagree, resolve it immediately rather than at the next annual review.

That final step is the one that catches transfer errors, and transfer is exactly when they occur. A schedule change is a manual operation somewhere in a billing system, and the first bill is where a wrong rate class, a missing rider or a carried-over determinant becomes visible. Checking it is an hour; discovering it eleven months later is a claim.

The model you built during [how to choose a rate schedule](https://thedemandcharge.com/articles/how-to-choose-a-rate-schedule) is what makes that check possible, which is a further reason to build it properly rather than accepting a vendor's comparison.


## Sources

- [National Association of Regulatory Utility Commissioners](https://www.naruc.org/)
- [California Public Utilities Commission](https://www.cpuc.ca.gov/)
- [Federal Energy Regulatory Commission](https://www.ferc.gov/)
- [Berkeley Lab — Energy Markets and Policy](https://emp.lbl.gov/)

## Frequently asked questions

### How long is a typical minimum stay?

Twelve months is common on optional schedules, and some carry longer commitments. The purpose is to prevent customers moving between schedules seasonally to take the best of each, which would defeat the rate design.

### Can I be moved to a different schedule without asking?

Yes, in some circumstances. Where a mandatory eligibility threshold is crossed — a demand level, a voltage level, a customer class definition — the utility may be required to reassign the account. Optional schedules require an election.

### Does switching reset a ratchet?

Not automatically, and this is worth confirming in writing before electing. The new schedule's ratchet clause may look back at demand recorded while you were on the old one.

### What if the switch turns out to be wrong?

If a minimum stay applies you generally serve it. That is precisely why the modeling should be tested against a changed load as well as a historical one before electing.

---

Canonical URL: https://thedemandcharge.com/articles/rate-switching-rules  
Site index for agents: https://thedemandcharge.com/llms.txt