Originally published in Climate Urgency, Climate Action
I keep telling everyone that we do measure development. In my mind, I have a pretty good idea of what that means. A client has a piece of tech, or a concept and they want a deemed savings value for it. We get their data, and everyone else’s data, and then make a bunch of “it depends” engineering decisions, plug it into the cost-effectiveness calculator, and ta-da! There’s a new measure with a deemed savings value that the utility or program can use to claim savings. Right?
I decided that I would ask our long-standing measure development expert, Jeremy Stapp, PE, about how we do it, what some of those tricky “it depends” questions are, and how he goes about the process.
Q: Jeremy, what exactly is measure development?
Measure development is a niche area within our industry. There are all sorts of different aspects of energy efficiency. Most people think of custom projects: you go to a site, install a widget, measure how much energy is being used, compare that to what was used before, and give a customer an incentive. Measure development is developing the calculations behind prescriptive measures so you can repeatedly install something and have a pretty good idea of how much energy it will save on average.
Will the measure, installed in potentially the thousands, actually save energy in the long term and be cost-effective?
We work with individual utilities or program administrators to develop measures. Utilities are obligated, usually by law, to offer measures that are cost effective, but that’s not the only limitation. Utility commissions also layer on additional requirements.
Q: How do you determine if a measure saves energy and is cost-effective?
When we do measure development work, we look at code requirements and equipment standards, as well as what the customer is already using, to understand the true counterfactual: what would have happened in the absence of the measure. When you’re designing the measure, you need to calculate based off a variety of different baseline scenarios. There are code baselines for new construction, and existing baselines for measures getting installed in existing buildings.
Then there’s also the question of what people are buying new if they don’t go through an efficiency program, but also, what are they buying used that we aren’t tracking in sales data?
Q: What kind of data goes into developing a measure?
We design the measure case by looking all around the industry at available alternative products, highest cost-effective efficiency levels, and potential market efficiency.
One thing to note is that we aren’t developing measure savings based off how much energy each installation will save: It’s not per project. Instead, we need to normalize to allow us to apply the savings to a wide range of use cases. For example, a motor will have savings calculated by horsepower. We have to go through the wide range of influencing factors and balance savings, normalization techniques, and market readiness.
It’s always a balance between convenience for scaling without sacrificing rigor.
Q: What do you do if there isn’t a good source for your data?
We pull all available studies, we vet the sources, the funding, the intent. For example, if a study is funded by the CEC (California Energy Commission), we might take it more seriously than one from a vendor, who may be more biased. But sometimes we’ll look at reputable organizations and see that they’ll have a whitepaper out that doesn’t have exact sources for their values. Then it’s a forensics exercise to dig deep and find the original sources and try to put together the preponderance of evidence that the sources justify the final value.
One time, we were looking at a controls measure where there was a new version of existing tech that had an automatic shutoff. The vendor produced a whitepaper, and the utility adopted the measure. The measure took off and it was the top measure in that state. But then, the third-party evaluation came along, and the realization rate ended up at 50%, and their cost effectiveness blew up. That measure had been a key measure for that program and the utility was left trying to plug a hole in their portfolio mid-cycle.
Q: How do you calculate incremental costs?
For each measure, we analyze the baseline cost. For new construction, this is based off the code or standard practice baseline. If it’s a retrofit of existing equipment, then that’s a more complicated statistical question about what exists in the market and at what level.
But in an early replacement scenario, which we lovingly call “everybody’s nightmare,” we have a dual baseline. We’ve got a motor in this example. It’s got plenty of life left in it, but it’s not very efficient. We need to then calculate the savings and costs for the remaining useful life of the original piece of equipment. The second baseline is the code (or market) baseline starting at the end of the original baseline. So you’ll have one incremental cost for the first baseline period, and one incremental cost for the second period.
Q: How do non-energy benefits factor into your work?
Each utility has a set portfolio of non-energy benefits and what dollar value is assigned to them. For water, let’s pretend the rate is $0.02 per gallon. We know how much water a particular irrigation measure will save, and we will build out the measure to normalize by gallons of water, and then be able to make an otherwise cost-ineffective measure pass the cost-effectiveness tests because the non-energy benefit of water saved is the clincher.
Q: How often do you update measures?
No measure is static, and we are constantly updating the baselines, the efficiencies, costs, etc., and adding in new data to ensure that the utility is getting a realistic representation of savings potential from their portfolios.