Farmers consume nearly 7% of California’s electricity. The vast majority of this electricity is used to power groundwater pumps, which farmers rely on for irrigating thirsty, high-value crops such as grapes, almonds, and pistachios. Groundwater is especially important during drought years, when farmers need to make up for surface water shortages by pulling water out of the Central Valley’s underground aquifers.
In a new Energy Institute working paper, Energy Institute alums Fiona Burlig, Louis Preonas, and Matt Woerman measure the extent to which higher electricity prices cause farmers to reduce their groundwater use.
The paper comes at a big moment in California policy—during the initial implementation of the Sustainable Groundwater Management Act (SGMA), the state’s first comprehensive groundwater regulation. Historically, groundwater in California had been treated like the wild west, with no pricing and no monitoring of groundwater usage leading to “a whole mess.” After over a century of unregulated groundwater pumping, California’s aquifers are “critically overdrafted”. SGMA requires local water management agencies to reduce groundwater extraction in order to achieve long-run sustainability targets. Given that the regions losing groundwater the fastest also produce California’s high-value perennial crops (see the figure below), SGMA will likely have major impacts on agriculture, water use, and electricity use.
—
—
Measuring how farmers respond to electricity prices
The authors take advantage of a unique natural experiment, which comes from the way Pacific Gas & Electric (PG&E) sets prices for its agricultural customers. Like households, farmers are able to choose from a menu of electricity tariffs, each with different combinations of fixed charges and price per kWh. However, unlike households, PG&E restricts each farmer’s menu based on the size of their groundwater pump (small vs. large) and their electricity meter type (conventional vs. smart meter). These restrictions lead otherwise similar farmers to face different prices over time.
—
—
Using this research design, the authors study electricity billing data to measure farmer’s demand responsiveness (many thanks to PG&E for making this work possible!). They find an annual elasticity of -0.76: farmers respond to a 10% increase in the price of electricity by reducing their electricity use by 7.6%.
In the context of electricity demand, this is a pretty big response—over 10 times the responsiveness of residential consumers in Sacramento. This suggests that farmers could be especially good targets for electricity demand response. In addition, the results suggest that even modest changes in pumping costs could have large consequences for California aquifers.
—
Short run vs. long run
The paper then goes on to distinguish between short- and long-run price responsiveness. The -0.76 elasticity is a “short-run” elasticity, measured using annual year-to-year price changes. The authors find that over a one-year period, very few farmers change crops, so this -0.76 elasticity reflects short-run water conservation behaviors.
Over the long-run, farmers have more options. For example, suppose that PG&E announces a permanent electricity price increase starting in 2027. In response to a permanent change like this, a farmer could decide to immediately stop planting new almond trees and instead change to a less-water intensive crop like tomatoes or to allow land to go fallow.
This short- vs. long-run distinction really matters, given that California agriculture is dominated by almonds, grapes, and other high-value perennial crops, which are planted once but bear fruit for many years.
Interestingly, when the authors use a dynamic discrete choice model to capture these long-run changes, they find a smaller elasticity of -0.37. This seems to violate Max’s Yoga Theorem: how could farmers be over twice as responsive in the short-run than the long-run, when the long-run is supposed to allow for more flexibility?
Fiona, Louis, and Matt argue (with strong supporting evidence) that it isn’t a question of flexibility, but of mechanisms. In the short run, farmers respond to price shocks using short-run strategies, like watering their crops (a lot) less throughout this growing season. These approaches are not sustainable in the long run, so farmers will look for more permanent solutions. Over longer time horizons, farmers consider switching crops: for example from perennial crops to less thirsty annual crops. Crop switching implies large changes in water consumption, but less overall than what is possible on a year-to-year basis.
—
The Sustainable Groundwater Management Act
The last part of the paper asks what this means for the future of California agriculture. SGMA will require roughly a 20% reduction in groundwater pumping, and the authors use their estimates to evaluate how this will affect crop choices. Here, a long-run perspective is crucial, since the policy will require more-or-less permanent reductions in pumping in order to achieve groundwater sustainability. Over half of California’s local management agencies are proposing price-based policies (such as “groundwater sustainability fees”). The evidence from changes in PG&E’s electricity prices is directly relevant—since higher electricity prices and groundwater fees (ahem, taxes) similarly raise farmers’ costs of pumping groundwater.
—
—
The authors predict that California’s sustainability goals can be achieved using the equivalent of a 60% tax on groundwater pumping—which would bring what farmers pay for groundwater more in line with the true costs of water consumption. The figure above shows that such a tax would have significant impacts on California’s agricultural sector, causing nearly 9% of cropland to switch categories. Fruit/nut perennial crops would decline by 24%, and fallowing (non-crop) would increase by 50% relative to a no-tax scenario. These results show that SGMA will seriously alter the landscape of crop production across California.
—
What does this all mean for electricity?
Meeting SGMA’s goals will also reduce electricity used for groundwater pumping by about 20%. Since groundwater pumping is responsible for about 7% of California’s electricity consumption, it follows that SGMA could reduce California’s electricity load by more than 1%. That’s in the neighborhood of 3 TWh per year—more than the generation of a typical combined-cycle natural gas plant, or enough to power half of the electric vehicles in California in recent years! Moreover, this reduction might be particularly helpful for California’s electricity grid, because pumping overwhelmingly takes place during the summer months, when demand for electricity is high.
We don’t tend to think about agriculture as being a big source of electricity demand, nor do we tend to think about groundwater regulations as being important for electricity markets. But this paper shows that California’s new landmark groundwater regulations have the potential to have exactly these kinds of large cross-cutting effects. Not only can SGMA help ensure that California’s aquifers are usable for years to come, but it also has the potential to significantly reduce the state’s electricity demand.
For more, see “Groundwater and Crop Choice in the Short and Long Run” (by Fiona Burlig, Louis Preonas, and Matt Woerman), Energy Institute Working Paper #349.
This blog is co-written with Lucas Davis from UCB.
In The News
