Latest / The Productivity Podcast with Fexingo: Output, Efficiency, and Long-Term Economic Growth / How a California Utility Solved the Duck Curve
Transcript
- Lucas: So there's this chart that energy economists love to show — it looks like a duck. The belly of the duck is midday, when solar is flooding the grid, and the neck is late afternoon, when the sun goes down and demand spikes just as solar drops off. That's the duck curve. Luna: And the duck has been getting fatter every year as more solar comes online. I've seen projections where the neck gets so steep you'd need gas plants to ramp up faster than they physically can. Lucas: Right. And that's a pure productivity problem — wasted energy, inefficient dispatch, and billions in stranded costs. But California has actually started to flatten the duck. And the story of how they did it is a really sharp case study in using market design to solve an infrastructure bottleneck. Luna: I want to hear the specifics. What was the first move? Lucas: Let me give you the lay of the land around 2020. California had installed something like 12 gigawatts of utility-scale solar, plus another 10 gigawatts of rooftop. By spring, when demand is low and sun is strong, you'd see midday net load drop to almost 10 gigawatts — compared to a peak of nearly 30 gigawatts in the evening. That's a 20-gigawatt swing in about four hours. Luna: And that swing is the duck neck. So what did they do? Batteries, obviously, but that's not the whole story. Lucas: Batteries are a huge part. But the really interesting piece is a market mechanism called the flexible ramping product — or FRP. CAISO, which runs the grid for most of California, introduced this around 2016, but it took a few years to mature. The FRP essentially pays generators and storage operators to be ready to ramp up or down on short notice. Luna: So it's a capacity payment for flexibility, not just energy. Lucas: Exactly. And the beauty is, it prices the option to ramp. Before FRP, the market only paid for actual energy delivered. So a gas plant that sat idle all day but needed to go from zero to full output in 15 minutes had no revenue stream to cover that readiness. FRP changed that. Luna: And that incentivized battery storage, because batteries are perfect for fast ramping. Lucas: Right. By 2025, California had installed over 10 gigawatts of battery storage — mostly four-hour lithium-ion systems. And they started to get dispatched in a really clever way: they charge during the solar belly, then discharge during the neck. That alone shaved about 4 gigawatts off the evening peak. Luna: But batteries alone don't fix the whole duck. What about the midday oversupply? Lucas: That's where time of use pricing comes in. In 2020, California had some of the highest solar curtailment rates in the country — about 6 percent of utility-scale solar was being turned off because there was nowhere to put the power. By 2025, that dropped to under 2 percent. Luna: How? Did they build more transmission? Lucas: Some, but the bigger driver was shifting demand into the middle of the day. The California Public Utilities Commission redesigned residential electricity rates so that the cheapest period is now 10 AM to 2 PM — when solar is peaking — and the most expensive is 4 PM to 9 PM. That pushed EV charging, water heating, and industrial processes into the solar hours. Luna: So you're basically using price signals to align consumption with generation. That's textbook economics, but getting it implemented politically is the hard part. Lucas: It was a fight. The initial proposal had a lot of pushback from consumer advocates who worried about shifting costs to low-income households. But they eventually settled on a three-tiered structure with a fixed low-income discount. And it worked. By 2024, midday demand was up about 8 percent compared to 2020, while peak evening demand grew only 1 percent. Luna: That's a measurable flattening. But I imagine the duck curve isn't just a California problem. Are other states copying this? Lucas: Hawaii was actually the first to really confront it, because they have so much solar relative to load. But California's model is being studied by ISO New England, ERCOT in Texas, and even parts of Australia. The flexible ramping product has been adopted by MISO in the Midwest and is being considered by PJM. Luna: So the productivity lesson here is that you can't just build more generation — you have to redesign the market to incentivize flexibility and demand shifting. Lucas: Exactly. And there's a broader point: infrastructure productivity isn't just about physical throughput. It's about how well the pricing and dispatch signals match the physics of the system. The duck curve is a case where better market design, combined with targeted technology deployment, avoided what could have been a crippling constraint on renewable growth. Luna: I want to get into the numbers on battery costs and how they fell, because that's part of the story too. Lucas: Sure. In 2020, battery storage cost about $350 per kilowatt-hour for a four-hour system. By 2025, that was down to around $150. That 60 percent drop in five years is what made the 10-gigawatt build-out economically viable. But even with that, the FRP and time of use rates were necessary to create the revenue streams. Luna: So it's a three-legged stool: cheap batteries, flexible market products, and dynamic pricing. Take any one away, and the duck doesn't get flattened. Lucas: And there's a fourth leg, actually: regulatory mandates. California passed a law in 2018 requiring all new homes to have solar panels, and in 2021 they started requiring solar plus storage for new commercial buildings. That forced a lot of behind the meter battery installation, which also helps. Luna: That's interesting. So the duck curve solution wasn't just one brilliant innovation — it was a coordinated set of policies and market tweaks that evolved over almost a decade. Lucas: Exactly. And that's the productivity story: improving the efficiency of the grid isn't about a single silver bullet. It's about aligning incentives, technology, and regulation so that the whole system operates closer to its potential. California's duck curve is still there, but it's much less duck-like than it was in 2020. Luna: Speaking of potential — if walking through the economy with us has made something click for you, we should mention that listener support is what keeps this podcast ad-free and independent. If today's episode was worth a coffee to you, there's a link at buy me a coffee dot com slash fexingo. Lucas: Yeah, it's a small ask, but it genuinely helps us keep making episodes like this one. No pressure at all. Luna: Alright, back to the duck. Lucas, you mentioned that the duck curve is flattening, but what's the next challenge? Lucas: The next challenge is the 'submarine curve' — that's what some analysts call the overnight low when wind and nuclear are still running but demand is minimal. As we add more wind, that becomes a new duck. But California's model gives us a template: flexible pricing, storage, and market products that reward flexibility. It's applicable everywhere. Luna: So the duck curve isn't dead, but it's been tamed — for now. Lucas: Tamed is the right word. And the productivity gain is real: less curtailment means more renewable energy actually gets used, which lowers the cost of decarbonization. That's a win for the economy and the grid.