💰 quantitative finance

Extended Version: Characterizing Distributed Photovoltaic Panel Investment Equilibria

This study models long-term distributed photovoltaic investment as a non-atomic game linked to short-term market equilibria, theoretically demonstrating that product-differentiated real-time markets achieve socially optimal capacity, single-product markets lead to under-investment, and contract-based markets can cause over-investment depending on user valuations.

Original authors: Mehdi Davoudi, Junjie Qin, Xiaojun Lin

Published 2026-02-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Mehdi Davoudi, Junjie Qin, Xiaojun Lin

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a giant, bustling marketplace where thousands of homeowners are deciding whether to put solar panels on their roofs. This paper is like a detective story trying to figure out: "How do the rules of the electricity market change how many people decide to buy solar panels in the long run?"

The authors, Mehdi Davoudi, Junjie Qin, and Xiaojun Lin, built a mathematical model to simulate this. They treat the market like a game where no single homeowner is big enough to move the needle alone, but the collective decision of everyone changes the price of electricity for everyone.

Here is the breakdown of their findings using simple analogies.

The Core Problem: The Feedback Loop

Think of the solar market as a two-way street:

The Short-Term Street: Every day, electricity prices go up and down based on how much sun is shining and how much power people are using. Homeowners sell their extra power here.
The Long-Term Street: Homeowners look at those daily prices and decide, "Is it worth spending $10,000 to install a panel?"

The catch? The daily prices depend on how many people already have panels. If everyone installs panels, the price of solar power drops (because there's too much of it). If no one installs them, the price stays high. The paper tries to find the "sweet spot" where the market settles into a stable balance.

The Three Market Rules (Mechanisms)

The authors tested three different ways the electricity market could be run to see which one leads to the best outcome for society.

1. The "Smoothie" Market (Single-Product Real-Time)

The Analogy: Imagine a coffee shop that sells "Caffeine." They mix regular coffee (grid power) and decaf (solar power) into one big pot. When you buy a cup, you don't know or care which bean it came from; it's just "Caffeine."
How it works: Solar energy is mixed with regular grid power. If you have solar panels, you sell your power into this big pot. You get the same price as the coal or gas power.
The Result: Under-investment.
- Why? Many people love solar because it's "green" and clean. In this "Smoothie" market, that green value is ignored. The price doesn't reflect the extra value people put on clean energy. So, fewer people install panels than society actually needs. It's like selling a premium organic apple in a bin with regular apples; you can't charge the extra price you deserve, so you stop planting organic trees.

2. The "VIP Section" Market (Product-Differentiated Real-Time)

The Analogy: Now, the coffee shop has two separate counters. One sells "Regular Caffeine" and the other sells "Organic Green Caffeine." People who care about the environment are willing to pay extra for the green cup.
How it works: Solar energy is treated as a special, distinct product. If you have solar panels, you sell directly to the "green" buyers who are willing to pay a premium.
The Result: Perfect Investment (Socially Optimal).
- Why? This market perfectly matches the supply of solar with the demand from people who value it. The price signals tell investors exactly how much solar is needed. It's the "Goldilocks" scenario—neither too much nor too little.

3. The "Future Contract" Market (Contract-Based)

The Analogy: Imagine a farmer selling a "subscription" to his future harvest before he even plants the seeds. He sells the right to 100 bushels of corn next year. The buyer pays now, hoping the corn will be good.
How it works: Instead of selling the actual electricity when the sun shines, panel owners sell "rights to capacity" (the promise of power) in advance.
The Result: Over-investment (usually).
- Why? When people don't care that much about the "green" aspect (the premium is small), this market gets a bit greedy. Investors bet that they can sell their capacity contracts easily, so they install too many panels.
- The Twist: The authors note this might actually be a good thing! If everyone installs too many panels, the cost of making them might drop (economies of scale), making solar cheaper for everyone in the future. It's a "build it and they will come" strategy that might lower costs long-term.

The Big Takeaways

If you mix everything together (Smoothie Market): You get too few solar panels. The market fails to reward the "green" value, so people don't invest enough.
If you separate the products (VIP Market): You get the perfect amount of solar panels. The market works exactly as it should.
If you sell future promises (Contract Market): You might get too many panels. But if the cost of making panels drops because of that over-production, it could actually help society in the long run.

Why This Matters

This paper is a blueprint for policymakers and utility companies. It tells them: "Don't just treat solar power like regular electricity. If you want to hit climate goals, you need market rules that recognize the special value of clean energy."

If we keep treating solar like a generic commodity (the Smoothie Market), we will under-invest and miss our green targets. But if we design markets that let solar shine as a unique product, we can reach the perfect balance of investment.

1. Problem Statement

The paper addresses the critical feedback loop between short-term electricity market equilibria and long-term distributed photovoltaic (PV) investment decisions.

Context: Individual investors (households and businesses) decide whether to install PV panels based on expected revenue over the panel's lifespan (e.g., 25 years).
The Feedback Loop:
1. Investment: Individual decisions aggregate to determine the total installed PV capacity ( $c$ ).
2. Market Impact: This aggregate capacity influences short-term distribution-level electricity markets (prices and allocations).
3. Revenue: Short-term market prices determine the revenue investors earn.
4. Equilibrium: Investors adjust their decisions based on these revenues until a Long-Term Nash Equilibrium (NE) is reached, where no investor has an incentive to deviate.
Core Question: How do different short-term market mechanisms (pricing and trading structures) shape the resulting long-term equilibrium aggregate PV capacity? Specifically, do these mechanisms lead to socially optimal investment, under-investment, or over-investment?

2. Methodology

The authors employ a rigorous mathematical framework combining game theory, optimization, and stochastic modeling.

A. Investment Model (Non-Atomic Game)

Players: A continuum of infinitesimal investors ( $I_{inv} := [0, 1]$ ), reflecting the reality of many small distributed generators with negligible individual market power.
Decision: Each investor $i$ chooses $x_i \in \{0, 1\}$ (install or not) for a fixed capacity $c$ .
Payoff: The payoff is the expected total revenue over the lifespan minus the upfront capital/installation cost ( $\pi_0$ ).
Equilibrium: A Nash Equilibrium (NE) is defined as an aggregate capacity $c$ where the marginal investor earns zero net profit (Revenue = Cost).

B. Unified Short-Term Market Framework

The authors develop a unified model to analyze three distinct market mechanisms, all modeled as competitive equilibria (CE) within a non-atomic game setting:

Single-Product Real-Time Market (srt): Solar energy and utility grid energy are pooled as a single product. Prices are determined ex-post based on realized supply and demand. Consumers cannot distinguish solar from grid power.
Product-Differentiated Real-Time Market (prt): Solar energy is treated as a distinct product. Consumers have a "solar premium" ( $v_i$ ) representing their extra willingness to pay for renewable energy. Prices are determined ex-post.
Contract-Based Panel Market (cb): Contracts are traded ex-ante (before generation is realized) based on panel capacity rather than realized energy. Buyers pay for the capacity to receive solar output, hedging against generation uncertainty.

C. Analytical Approach

Derivation: The authors derive closed-form expressions for short-term market equilibria (prices and allocations) for each mechanism.
Linkage: These short-term revenues are fed into the investment model to derive the condition for the long-term NE capacity.
Benchmarking: A Social Welfare Optimal (SWO) capacity ( $c_{opt}$ ) is defined as the capacity that maximizes the sum of consumer and investor payoffs.
Comparison: The derived NE capacities are compared analytically and numerically against $c_{opt}$ under varying conditions (e.g., magnitude of consumer solar premiums, capital costs).
Extensions: The model is extended to handle heterogeneous operation periods (e.g., seasonal or daily variations in load and solar irradiation).

3. Key Contributions

Analytical Tractability: The paper provides a novel, analytically tractable framework linking short-term market equilibria to long-term investment equilibria using non-atomic games. This avoids the need for purely numerical simulations common in prior literature.
Unified Market Modeling: It introduces a general framework capable of representing diverse mechanisms (real-time vs. contract-based, single-product vs. differentiated) with closed-form solutions.
Theoretical Characterization of Market Failures:
- Single-Product Market: Proves that treating solar and grid energy as identical leads to under-investment relative to the social optimum because it fails to capture the value consumers place on renewable attributes.
- Product-Differentiated Market: Proves that distinguishing solar energy leads to the socially optimal investment level ( $c_{ne}^{prt} = c_{opt}$ ) because the market price internalizes the consumer's premium for green energy.
- Contract-Based Market: Demonstrates that trading capacity ex-ante can lead to over-investment ( $c_{ne}^{cb} > c_{opt}$ ), particularly when consumer solar premiums are small. This occurs because the contract mechanism decouples the payment from the realized scarcity value of solar energy.

4. Key Results

Theoretical Findings

Inequality of Equilibria: Generally, $c_{ne}^{srt} \leq c_{ne}^{prt} = c_{opt}$ . The single-product market always results in lower capacity than the socially optimal level (unless consumers value solar and grid energy identically, i.e., premium = 0).
Over-Investment in Contracts: When consumers have a small willingness to pay for solar (small $\epsilon$ ), the contract-based market leads to over-investment ( $c_{ne}^{cb} > c_{opt}$ ).
Zero Premium Case: If consumers place no extra value on solar ( $v_i = 0$ ), all three mechanisms yield the same equilibrium capacity, which equals the social optimum.

Numerical Validation

Using data from California (solar irradiation, consumer willingness-to-pay, and utility rates):

Base Case: With realistic solar premiums, the single-product market results in significantly lower capacity (67.88 GW) compared to the social optimum (70.42 GW).
Contract Market: The contract-based market consistently yields the highest capacity (71.57 GW), confirming the over-investment trend.
Sensitivity: As consumer solar premiums increase, the gap between the single-product market and the social optimum widens, while the contract-based market continues to over-invest.
Cost Sensitivity: Increasing capital costs reduces capacity across all markets, but the single-product market is the first to become unprofitable (capacity drops to zero) at high cost thresholds.

5. Significance and Implications

Policy Design: The study provides strong theoretical evidence that product differentiation (e.g., green tariffs, renewable energy certificates integrated into real-time pricing) is crucial for achieving socially optimal solar adoption. Relying on standard single-product markets may lead to sub-optimal renewable penetration.
Market Mechanism Trade-offs: While contract-based markets might seem attractive for stability, they risk distorting investment signals, potentially leading to inefficient over-capacity unless economies of scale (which lower future costs) are factored into the dynamic analysis.
Planning: Utilities and Independent System Operators (ISOs) must account for the specific market design when forecasting long-term distributed generation growth, as the mechanism itself dictates the equilibrium capacity.
Methodological Advance: The use of non-atomic games allows for precise analytical comparisons that were previously only possible through simulation, offering a robust tool for future market design research.

In summary, the paper establishes that market design is not just a short-term operational detail but a fundamental driver of long-term renewable energy investment, with product-differentiated markets emerging as the theoretically superior mechanism for social welfare.