19 Oct Why do we still let electric utilities write rates in words? Efficient markets demand clear pricing signals.
I try to avoid posts that proclaim how the world would be a better place if only… blah, blah, blah (fill in your favorite panacea here). But I’m going to indulge here only because I see our industry trying to adapt to disruptive trends, yet hamstrung with outdated means of communicating price signals to consumers. I see a big incongruity between what we’re trying to do with modern approaches to reimagining the electric grid and the current realities of electric rate tariffs.
Verbose utility rate language hinders optimal and efficient energy spending
My premise is simple; if we’re to accomplish the significant load-shaping that we need to do to bring about the grid many are imagining, we’re going to need much more clear price signals for consumers to plan around. And the fact that we allow rates to be written as legal fillings – in words – is at odds with how we need to communicate pricing.
Our current rate schedules are typically difficult to understand, estimate energy costs, and analyze to quantify the impact of demand response choices.
Rates have always been too complicated for most consumers to understand, so added simplicity and predictability would be great. But if we’re transitioning to a world where consumers must make real-time or day-ahead decisions about when to use, produce, shed, store or discharge stored electricity, getting clear and consistent price signals is going to be key.
As I mentioned a couple of week’s ago, in the new grid, timing is everything. We need rates that are simple, and perhaps more importantly, are easy to analyze on the fly. If I’m a sophisticated customer in the new world grid, every hour I have the potential to optimize my utility spend. The optimization problem can get tricky quickly.
What’s driving up my demand charges?
Rates are so complicated now that even sophisticated consumers often don’t understand what they are paying, and for which kilowatt hours. Many of our customers have seen a big rise in demand charges so that demand can now be 50% of your monthly bill. That means that, at a minimum, half of your electric bill depends on the highest 15-minute consumption of the month. So, you better be paying attention to what you’re doing in those 15 minutes.
It can be even more extreme if you have a demand ratchet. Under those tariffs your highest 15-minute demand can impact your charges for a year. But you might have a hard time figuring that out if you read the fine print of your rate. Those documents tend to be rife with legal-ease, confusing terms, complicated seasonal and time-of-use (TOU) charges. As long as we let attorneys write those rates in words, they can be unnecessarily difficult to understand, and predict what you’ll actually pay.
Energy storage and load shifting. Simple, no?
Let’s look at the (relatively) simple case where I can store energy in a battery to shift load. My control sequence for using that battery needs to be able to value every kWh of electricity just as the utility will. If I don’t value an August day for the annual peak that it sets on my ratchet, and simply give it the commodity time of day price for those peak hours, I could grossly underestimate the impact on my bill, and future bills, for that peak.
Of course, energy storage folks will tell you that they have all that dialed in, but I’m skeptical. Having spent a career digging around controls and energy issues in buildings, I’ve found few controls approaches that worked correctly out of the gate. I don’t see any reason why this burgeoning industry would be any different.
What impacts your electricity bill?
Have you ever tried to re-create your utility bill? For most commercial customers it could literally take hours for a given month. Here are some of the obfuscating factors:
- Time-of-day peak demand charges (often for 3 periods per day)
- Time of day consumption (i.e. commodity) charges (ditto)
- Seasonal changes in demand and consumption charges (typically multiply above by for winter/summer)
- Multiple pricing tiers depending on service voltage
- Baseline level of consumption (tiers or blocks)
- Peak demand ratchets (e.g. annual peak charges)
- Power quality charges
- Assorted add-ons like special levies, taxes, capital recovery mechanisms, etc.
This complication has led to a cottage industry of experts who perform “audits” on bills to make sure that 1) you’re on the right rate and 2) are getting charged correctly on that rate. The Houston Chronicle reported this summer that:
“At least five companies in Texas are providing both free and paid services aimed at helping consumers in Houston and other deregulated markets decipher confusing electricity offers such as free nights and weekends, multitiered pricing plans, and credits for high electricity use.”
And this is just in the residential market. Those are supposed to be the simple rates.
Rate complications have encouraged many billing analysis firms to develop “rate engines” that can simulate the on-the-bill impact of your energy use matched with various tariffs (our friends at Gridium do a nice job of this btw). But setting an algorithm to respond to your current rate with your current equipment is actually a much tougher optimization problem.
How should we define utility rates if not in words?
It’s always easier to complain about the current state of affairs than to offer an alternative so let’s consider a couple of potential options. Ultimately it will be up to each state’s electric regulators (Public Utility Commissions and Public Service Commissions) to do something about it as they typically have the final say on rates proposed by utilities.
Sure, engineers like us love a good optimization puzzle to solve, right? But on an hourly basis, not so much. We need software and algorithms to smartly adapt to price signals, some of which may only know a day, or hours before a peak event. That’s why we need rates that are delivered in standardized formats that are easy to analyze.
One simplistic example of an electric rate offered as an algorithm instead of written in words might be a simple array of kWh charges by hour for each of the 8760 hours in a year. A little clunky perhaps but easy to interpret and employ when I’m making decisions. Back during deregulation in California around 1999, we had some customers getting essentially that on real-time pricing rates. They would get a FAX (ha!) the night before with the commodity price for the next 24 hours.
A handier option might be a simple Python routine that returns the expected cost for an hour depending on the relevant variables. In other words, instead of getting a tariff in words, I’d get a math function that told me my expected costs for a given hour. For example,
cost[per kWh] = function(date, hour, expected kW, expected kWh)
That code would then be the basis for my optimization on a daily, or hourly basis.
Clearly there are charge types that are harder to encapsulate this way (e.g. ratchets again). But those could be treated by making the prior peak kW (annual or month) an input to the rate function. That approach would make the real impact of that hour show up with the true cost that you will bare for increasing the service the utility is obliged to provide. The other alternative is to move that discussion outside of hourly rates, and into the commodity negotiation in the first place.
What is the path to clearly, understood utility rates for optimal efficiency?
What are your thoughts on this approach? I’d love to hear what you think about the state of utility rate schedules and potential solutions. It does seem like we’re letting the utilities off the hook too easy on this one. When you grant a monopoly of service, you can ask for special considerations as part of that agreement. It seems asking for easy-to-understand, easy to quantify, and easy-to-automate rates ought to be part of that bargain when we’re paying for it at the end of the day. If you need help considering your next demand side management project or understanding your utility bill, we’re here to help anytime.