By: Stephen Shiwei Wang and Diana Jinghan Zhou
Edited By: Sayidcali Ahmed
Introduction
A major issue in the current energy transition is that the derived and induced demand for electricity from perceived alternative energy sources is growing faster than the deployment of renewable energy, creating a demand-supply gap that negatively impacts the transition.
Global energy consumption has been growing annually for more than half a century.1 The US electricity sector comprises power plants that generate at least one megawatt of electricity and serve end users, including the transportation, industrial, residential, and commercial sectors.2 The US generates more electricity each year. In 2024, US total energy production reached a record level, exceeding 103 quadrillion British Thermal Units.3 On the other hand, U.S. electricity consumption in 2024 reached a record level, driven by the development of data centers and the industrial sector; the U.S. Energy Information Administration (EIA) projects electricity use to grow by 1.7 percent each year through 2026.4
Renewable energy sources are expanding rapidly year by year, but fossil fuels remain dominant. Natural gas accounted for 38 percent of US total energy production in 2024, and crude oil 27 percent.5 Even as renewable energy production reaches record levels, fossil fuels still seem to dominate.
The process for end users to use electricity involves a chain of operations: energy is stored in raw natural resources or mechanical movements on Earth; then these potentials are converted into heat sources or mechanical movements to drive an electricity generator; the electricity will then go through two distinct processes, transmission and distribution, after which the end users can use it. The fundamentals of electricity generation are electromagnetic induction, which happens when a conductor “cuts through” a magnetic field.6 The raw inputs and natural potentials are used to power this movement for electricity generation.
As both electricity demand and supply grow, an imbalance of market equilibrium arises. We propose a framework that classifies the forces shaping the U.S. electricity sector along two dimensions: 1) demand growth, and 2) factors of electricity supply. Demand growth has two parts: derived demand resulting from increases in non-negotiable electricity use driven by technological innovations and implementation, and induced demand from behavioral responses to the perceived increase in renewable energy. Factors of supply include: the expansion of renewable energies, and delivery of constraints in transmission, distribution, and grid capacity. We then mapped these factors onto a two-by-two matrix to visualize the distinct mechanisms arising from different combinations of demand supply scenarios. This framework provides a structural explanation for today’s hold up in the energy transition despite renewable expansions.
Derived Demand
Derived demand is the need for a factor of production or an intermediate good. For example, industrial energy demand is a derived demand resulting from a firm’s output. In other words, electricity is not sought for its own sake but for the functions it provides. The electrification of transportation, heating, industrial automation, and digital services all increases electricity demand. The rise of artificial intelligence and electric vehicles has also driven demand for data centers and charging station infrastructure. These systems operate only with a continuous, stable, and high quality power supply, and a typical data center campus consumes “the same amount of power and water as an entire city.”7 Once these infrastructures are established, their electricity demand becomes inelastic due to ongoing maintenance and the computing services they support, such as powering AI informational services. As a result, technological and infrastructural advancements have led to increased electricity demand.
Induced Demand
Electricity consumption also depends on the perception of supply. Induced demand first appeared in transportation research, and it explained why simply adding lanes to highways cannot solve traffic congestion. Induced demand starts from an expanded capacity provided and then demand reacts to the sudden increase in quantity supplied. Neoclassical studies of economics tend to assume that demand is fixed in the short run, so induced demand is a change in demand volume, seen as a movement along the demand curve.8 Or precisely, an endogenous shift in the short-run of demand curve.
Induced demand also exists in the energy sector. For residential electricity use, if households perceive electricity is 1) cleaner, so there is less impact to the environment, and 2) cheaper, because the composition of renewable electricity is growing, then this perception can reduce conservation efforts, and in turn encourage an increased use of appliances. In case 1), demand is an outcome of social practice.9 The practice of conserving energy stems from reducing carbon emissions, and as renewables take a greater share of electricity generation, existing practices- guilt- will change. In behavioral terms, this can be a case of moral self-licensing.10 Research also demonstrated that installing solar panels reduces household dependence on the grid, and at the same time, the decrease is much smaller than the increase in electricity consumption.11 Induced demand reflects changes in behavior because of a change in information and expectations.
Supply: Renewable Capacity
According to the EIA report, renewable energy generation has grown substantially in recent years. Renewable capacity refers to the amount of power, in theory, that can be harnessed, converted, and generated into electricity by existing technologies. Renewables, including wind, solar, geothermal, and water, all have corresponding technologies for electricity generation. The expansion of renewables capacity supports long term grid development and decarbonization, but it is independent of electricity reliance on fossil fuels and does not directly ensure electricity availability.
Supply: Deliverability of Electricity
The electricity delivery system encompasses the processes of electricity transmission and distribution, as well as the relevant technicalities and infrastructure issues. First, electricity deliverability is often constrained by grid congestion because rising demand doesn’t match existing grid capacity. Second, the grid may be subject to intermittent electricity supply, especially with the increase in renewables. With fossil fuels consistently generating high heat for electricity generation, renewable energy sources such as wind and solar can be intermittent and unpredictable, depending on a more complex system, the weather.12 As a result, a large volume of renewable energy is curtailed, posing “a significant challenge to the efficient utilization of sustainable energy sources”, and the curtailment leads to increased carbon emissions.13 The reliability of electricity delivery can help explain the continued reliance on fossil fuels.
The Demand-Supply Framework
The issues of electricity usage, consumptive behaviors, technology constraints, and delivery are often discussed separately, and rarely organized and framed as a single analytical model. Using a matrix, the two types of demand and supply can illustrate four different scenarios as states of energy transition, and it is a straightforward way to analyze how various forms of electricity consumption interact with different supply factors.
|
|
Derived Demand |
Induced Demand |
|
Renewable Capacity |
I: Productive Use of Renewables/Futures of Energy |
II: Behavioral Rebound |
|
Deliverability Constraints |
III: Engineering Challenges/Grid Design Issues |
IV: Energy Use Crowding Out & Behavioral Rebound Externality |
Quadrant I: Productive Use of Renewables
When non-negotiable, inelastic demand for electricity relies on expanded renewable capacity, this represents the future of the energy industry. In this quadrant, electricity generated by renewable energy is directly consumed by large demanders, such as data centers, EV infrastructure, and electrified industrial factories. This requires renewable capacity to scale up to meet this high baseline electricity demand. When these are aligned, renewables will dominate, allowing a natural phase out of fossil fuels.
Quadrant II: Behavioral Rebound
This quadrant illustrates the interaction between induced demand and the growth of renewable energy capacity. It highlights the socio psychological responses to the expansion of renewables, such as solar rebound research, moral self-licensing, and increased consumption driven by shifts in practice and social norms. As renewable energy capacity rises, it unintentionally encourages additional electricity use yet still signifies progress in decarbonization. This scenario emphasizes the internal behavioral responses of residential and commercial users to the growing abundance of electricity.
Quadrant III: Engineering Challenges and Grid Issues
As the increase in derived demand meets the physical constraints of the grid, it becomes a technical issue. This demand is driven by technological advancements that require electricity. This stage involves transmission and distribution, grid congestion, and stability problems. As derived demand rises, the availability of electricity is limited.
Quadrant IV: Energy Use Crowding Out
The deliverability issues from the additional use of electricity from induced demand are its negative externalities. This is evident through rises in electricity consumption that lead to grid congestion and peak-load stress. It represents a behavioral increase in electricity use, thereby straining the technical aspects of electricity delivery. This is a classic crowding-out problem: the reason residential and commercial users increase electricity consumption is due to the expansion of renewables, but this behavior adds extra stress to the electricity distribution networks. These stresses translate into electricity rate hikes, as reflected by regional Independent System Operators (ISOs), or through delayed electrification plans and fossil fuel phase-outs.
Applying the Framework to Smart Grid Technologies
A smart grid aims to address the fast-growing utility demands of American homes and businesses, manage variable demand load patterns, and improve grid stability. First, rising total electricity use reflects an increase in derived demand; second, maintaining system reliability under these expanding, uneven loads poses a deliverability problem. Together, these two problems directly map to quadrant III: engineering challenges and grid issues. As demand rises especially during peak-and-tough periods, the grid can experience congestion, leading to delivery issues. Smart grids serve as tools for modeling traffic and energy storage, offering solutions to these delivery constraints. However, by improving reliability and capacity, a newly implemented smart grid could also shift the system toward Quadrant II: Behavioral Rebound, where users may expand their consumption in response to perceived abundance.
Conclusion: Defining the Framework
Given an electricity system with clearly defined demand drivers and supply scopes, this framework locates the system within a four-quadrant grid, analyzes the structural forces, and provides potential system evolutions. This diagnostic tool helps organize the dynamics of the fast-paced renewable energy development and electricity use evolution, and how solving one problem leads the system into a new state of development.
At the coarsest level, if each quadrant is treated as a “regime” and each regime is further broken down into discrete states based on the magnitude of demand and supply, a Markov model can efficiently estimate the probabilities of moving from one state to another. This modeling could help identify dominant pressures within the system at each state. But at the same time the model is prone to diverge under unforeseen policy, economic, or technological shocks.
Works Cited
- Hannah Ritchie, Pablo Rosado, and Max Roser. 2020. “Energy Production and Consumption” Our World In Data. Last updated January 2024. https://archive.ourworldindata.org/20251125-173858/energy-production-consumption.html.
- U.S. Energy Information Administration. 2024. “U.S. Energy Facts Explained”. Last updated July 15. https://www.eia.gov/energyexplained/us-energy-facts/.
- U.S. Energy Information Administration. 2025. “In 2024, the United States Produced More Energy Than Ever Before”. Published June 9. Principal contributor: Mickey Francis. https://www.eia.gov/todayinenergy/detail.php?id=65445.
- U.S. Energy Information Administration. 2025. “After more than a decade of little change, U.S. electricity consumption is rising again”. Published May 13. Principal contributors: Mark Schipper, Tyler Hodge. https://www.eia.gov/todayinenergy/detail.php?id=65264.
- U.S. Energy Information Administration. 2025. “In 2024, the United States Produced More Energy Than Ever Before”. Published June 9. Principal contributor: Mickey Francis. https://www.eia.gov/todayinenergy/detail.php?id=65445.
- U.S. Energy Information Administration. 2023. “Electricity Explained. How Electricity Is Generated.” Last updated: October 31. https://www.eia.gov/energyexplained/electricity/how-electricity-is-generated.php.
- Dayarathna, Miyuru, Yonggang Wen, and Rui Fan. 2016. “Data Center Energy Consumption Modeling: A Survey.” IEEE Communications Surveys & Tutorials 18. No. 1: 732–94. https://doi.org/10.1109/COMST.2015.2481183.
- Lee Jr, Douglass B., Lisa A. Klein, and Gregorio Camus. 1999. “Induced traffic and induced demand.” Transportation Research Record 1659, no. 1 (1999): 68-75.
- Rinkinen, Jenny, Elizabeth Shove, and Greg Marsden. 2020. Conceptualising Demand: A Distinctive Approach to Consumption and Practice. London: Routledge. https://doi.org/10.4324/9781003029113.
- Lalot, Fanny, Juan Manuel Falomir-Pichastor, and Alain Quiamzade. 2022. “Regulatory Focus and Self-Licensing Dynamics: A Motivational Account of Behavioural Consistency and Balancing.” Journal of Environmental Psychology 79: 101731. https://doi.org/10.1016/j.jenvp.2021.101731.
- Nguyen, Luan Thanh, Shyama Ratnasiri, Liam Wagner, Dan The Nguyen, and Nicholas Rohde. 2024. “Solar Rebound Effects: Short and Long Term Dynamics.” Renewable Energy 223: 120051. https://doi.org/10.1016/j.renene.2024.120051.
- Gross, Samantha. 2020. “Why Are Fossil Fuels So Hard to Quit?” Brookings Institution, June 2020. https://www.brookings.edu/articles/why-are-fossil-fuels-so-hard-to-quit/.
- Laimon, M. 2025. “Renewable Energy Curtailment: A Problem or an Opportunity?” Results in Engineering 26: 104925. https://doi.org/10.1016/j.rineng.2025.104925.
Author Bio
Stephen Shiwei Wang is a second-year MPA candidate with a concentration in Economic & Financial Policy and the EFII fellow. He aims to bring policy and regulation analysis frameworks to revolutionize the financial services industry, particularly in risk management. Stephen has a strong interest in behavioral economics and econometrics, with extensive research experience with STATA and QGIS. He also enjoys studying Logic in Philosophy and believes it often provides a more robust explanation of the world than most traditional economic theories. In the short term, Stephen hopes to combine his policy analysis training with his financial analysis expertise. In the long term, Stephen aspires to become an activist investor in Asian emerging markets to accelerate social development.
Diana Jinghan Zhou is an undergraduate student at the University of Edinburgh, currently pursuing an MA in Economics. She has gained extensive industry experience across banking and securities through internships focused on corporate financing, debt instruments, and M&A deal structuring. Her interests center on growth capital, particularly innovative financial products that address wealth-redistribution challenges, and technologies with strong scaling potential in emerging markets — those tackling global logistics bottlenecks, labor-market frictions, and the decentralization of production networks in the global south. Diana has also contributed to research on the effectiveness of China’s poverty-alleviation policies and conducted a comparative study on U.S. and EU place-based policy development strategies.
