After years of flat or even declining growth, electricity load forecasts are skyrocketing. It’s not difficult to find headlines focused on the coming electricity surge — and the implications thereof. PJM Interconnection, for example, recently tripled its yearly load growth forecast. And nationwide, data compiled by the Federal Energy Regulatory Commission (FERC) shows that load growth forecasts for the next five years have jumped over 80%.
The growth in data centers required to power the artificial intelligence (AI) boom only complicates the picture. It currently takes 10 times as much electricity to run a ChatGPT query compared to a traditional Google search. As AI gets more sophisticated, those power demands may even grow: AI data centers could consume up to 12% of the nation’s electricity by 2028.
This load growth is raising concerns about reliability and affordability. Whether our grid can even handle the rapidly increasing load demands —from AI but also electrification of buildings and vehicles and onshoring of industrial processes — is now being debated. The good news is we can handle the power demands of the 21st century, but only if we develop smart policies that properly value energy efficiency and other “demand flexibility” resources.
A recent report released by Berkeley Lab summarizes well how energy efficiency can contribute to our growing energy needs: “Prioritizing investments aligned with grid needs allows for energy efficiency to provide value beyond annual energy savings, such as helping utilities meet peak demands or relieving transmission and distribution system constraints.”
The cleanest, cheapest kilowatt is the one not used, which makes energy efficiency the least expensive resource we have. No less an authority than the International Energy Agency (IEA) has pegged energy efficiency as the single-largest measure to avoid energy demand in the net-zero emissions by 2050 scenario. Energy efficiency can help us avoid building more supply-side infrastructure, reduce costs, and eliminate energy waste, all of which are crucial for improving affordability and grid reliability.
Indeed, energy efficiency is the least expensive resource we have, one that will minimize the buildout of expensive generation, transmission, and distribution. But to play a crucial role in meeting our growing energy needs, energy efficiency needs to evolve and get real — literally and figuratively.
It’s time for energy efficiency to get real
Currently, most energy efficiency programs are built to meet compliance goals. Policymakers prescribe budgets and/or energy savings goals and task utilities or government agencies to implement energy efficiency programs that meet these goals. In general, policymakers want these programs to contribute to broader energy policy goals and ask the program administrators to quantify the impacts of these public investments in terms of energy savings, bill savings, environmental impacts, and other metrics.
Today, most of these compliance-based efficiency programs typically calculate impacts via “Deemed” or “Modeled” methods:
- Deemed energy efficiency programs use predetermined savings values based on typical savings for specific efficiency measures.
- Modeled energy efficiency programs use simulation models to predict the energy savings of an efficiency improvement. These models are often customized for each project based on site-specific details and operating conditions.
The challenge is that these methods to calculate program impacts typically fail to align with the standards of accuracy and/or granularity required for reliability and resource adequacy purposes. In general, grid operators — be they utilities or Regional Transmission Organizations (RTOs) like PJM, NYISO, ISO-New England, MISO, or CAISO — require that resources be measured at the meter across time, location, and other variables.
When those standards are not met, grid operators make clear they will not incorporate the impacts of energy efficiency. For example, FERC has approved PJM Interconnection’s plan to eliminate energy efficiency from the capacity market, in part because of concerns about the reliability of energy efficiency. PJM argues that energy efficiency lacks the same level of verifiability as traditional supply-side resources, raising doubts about its ability to contribute consistently to grid reliability. In a criticism of past capacity market rules, PJM stated that they “do not explicitly require a showing that capacity payments to Energy Efficiency Resources actually effectuate any reduction in energy consumption…” In short, it is hard to value energy efficiency like a supply-side resource when you aren’t confident in how much energy is being reduced.
Reliance on the compliance framework is also starting to raise issues in states with a long history of supporting investments in energy efficiency. In Massachusetts, often listed as one of the top states for energy efficiency by the American Council for an Energy Efficient Economy (ACEEE), policymakers and advocates are raising questions about whether the large energy efficiency investments the state makes are actually achieving the desired results. The Department of Public Utilities (DPU) recently slashed the Mass Save budget by $500 million, stating “[t]he Department…is cautious about increasing ratepayer burdens when the benefits of the energy efficiency programs, particularly those focused on decarbonization, may not be evident to customers through lower bills”. In other words, the DPU is struggling to confidently translate the savings from energy efficiency to bills.
California, another top ACEEE state, faces a different kind of challenge to make energy efficiency a real resource. Californians endure some of the highest energy prices in the country, leading to Governor Newsom filing a recent Executive Order to drive greater energy affordability. In response, the California Public Utility Commission (CPUC) as well as the California State Auditor has been asked to review the “cost-effectiveness” of ratepayer-supported energy efficiency programs (among other things). The problem is that the CPUC cost-effectiveness framework doesn’t actually treat energy efficiency as a resource, instead using a Total Resource Cost (TRC) test that penalizes private capital for deep retrofits that drive the most ratepayer value. As the recent report from the California State Auditor stated:
“The approach the CPUC takes to measure cost-effectiveness with its calculation of the TRC may discourage utilities from implementing certain efficiency programs, and the approach may contribute to utilities regularly not having cost-effective program portfolios.”
In other words, California (among other states) isn’t even defining cost-effectiveness in terms of real ratepayer value, leading to a significant underinvestment in energy efficiency as a system resource.
When Massachusetts and California spend roughly $2.5+ billion per year on energy efficiency without confidently creating ratepayer value, not only does it needlessly raise questions about whether paper savings are real savings, but it also makes it harder to demonstrate that energy efficiency is a worthwhile investment, especially when ratepayer funds are involved. For energy efficiency to be counted as a grid resource, it therefore needs to be considered reliable, real, and beneficial by the people who are accountable for resource adequacy and ratepayer benefits.
Energy efficiency can be a resource today with measured savings
Making energy efficiency a real, reliable (and measured) resource can unlock the full potential of Virtual Power Plants (VPPs) to make America’s grids reliable, affordable, and sustainable. VPPs can be employed to optimize electricity demand and lower overall grid costs. And VPPs have huge potential. The U.S. Department of Energy estimates that $10+ billion in grid costs can be saved annually by scaling up these virtual plants.
States can make energy efficiency a real resource today by passing laws and regulatory orders that create measured savings programs, and ensure that these savings are properly valued in the context of cost-effectiveness systems focused on system and ratepayer benefits. The good news is that some states are already doing this (and it’s often not the ones you think). Both Kansas and Louisiana, for example, have established a preference for energy efficiency programs to be measured. Frameworks like Total System Benefits (adopted by California, for example), and tools like the National Standard Practice Manual are providing ways for states (like Maryland and Michigan) to move beyond out-dated cost-effectiveness frameworks.
One of the best, near-term opportunities is for states to shift existing energy efficiency budgets to measured savings programs. States including California, New Jersey, and New York have started doing this — but more states need to shift towards measured savings if we are going to truly tackle load growth challenges. Additionally, states should opt to offer the measured approach when rolling out the Inflation Reduction Act HOMES Program — which many states from California to Florida have opted to do.
The forthcoming load growth is real — and this makes energy efficiency more important than ever. But we have to optimize its full potential and value, and that requires moving as much as possible toward measured savings programs that incentivize verifiable energy reductions. We can’t afford energy efficiency programs that meet noble compliance goals but do not tangibly (and measurably) contribute to our energy system. Ensuring that energy efficiency gets real is the way forward — for affordability, grid reliability, and competitiveness as electricity growth ramps up in the years to come.