Vertical Integration or Decentralization in the Power Sector

Lighthouse Energy Institute. American Enterprise Institute. September 23, 2026. Presenter: Devin Hartman.

Who We Are

LEI is an independent, nonpartisan think tank. Our mission is to provide information that improves the quality of energy and environmental institutions and policies. LEI achieves this by conducting honest, robust research and education consistent with three principles: competitive markets, limited government, and quality institutions.

Facilitating Competition Requires Vertical Unbundling

Cost-of-service (CoS): regulation substitutes for competition

• Natural monopoly premise: supply and delivery
• Goal: correct for perverse utility incentives (boost rate base, RoR)
• Reality: major economic discipline problems
  • Moral hazard: high costs, stranded assets
• Consumers and economists call for competitive reforms

Restructured: regulation facilitates competition

• Necessitates vertical unbundling
• Separate competitive supply from CoS delivery
  • Lose supply-delivery synergies
• Parallels to, but more complex than, telecom, natural gas, railroad liberalization
• Supply: generation and retail components

Merits: does competitive supply outweigh the side effects of unbundling?

Sources: Joskow (2006), Hogan (2010), Lien (2008)

Structural Overview

Diagram: A vertically integrated power monopoly runs Generation, Transmission, Distribution, and Retail as a single chain. In an unbundled electricity market, generation (G1, G2 … Gn) and retail (R1, R2 … Rn) are market activities, while transmission and distribution remain regulated activities.

Map of U.S. states by market structure:

• Traditional: Washington, Oregon, Idaho, Montana, Wyoming, Nevada, Utah, Colorado, Arizona, New Mexico, Alaska, Hawaii, Tennessee, North Carolina, South Carolina, Georgia, Alabama, Florida
• Hybrid: California, North Dakota, South Dakota, Nebraska, Kansas, Oklahoma, Minnesota, Iowa, Missouri, Arkansas, Louisiana, Mississippi, Wisconsin, Michigan, Indiana, Kentucky, West Virginia, Virginia, Vermont
• Restructured: Texas, Illinois, Ohio, Pennsylvania, New York, Maine, New Hampshire, Massachusetts, Rhode Island, Connecticut, New Jersey, Delaware, Maryland, District of Columbia

Note: participation in an organized energy imbalance market is not considered an RTO. California, Michigan and Virginia are in RTOs with a significant minority market share for competitive generators and limited access to competitive retail providers. Georgia and Oregon allow limited direct market access options for large customers, but do not have an RTO.

RTO/ISOs facilitate wholesale competition:

• Independent market administration required for restructuring
• Optional for integrated utilities (hybrid vs. traditional)

Sources: Arocena & Kumbhakar (2021), Giberson & Hartman (2023)

Restructuring Effects

Beneficial: Supply cost; Supply risk; Demand participation; Innovation; Environment; Transparency; Electoral governance; Business customer welfare; Customer self-supply.

Inconclusive: Transmission and distribution; Interconnection; Reliability; Market power and manipulation; Grid governance; Residential customer welfare.

Harmful: none identified.

Sources: Bushnell et al. (2017), Giberson & Hartman (2023)

Investment Supply

Core restructuring thesis, under-examined:

• Shifted incentives, locus of decisions
• Causal economic efficiency evidence (2025 study)
• CoS distorts investment decisions (2025 study)

2000s: unrealized load growth, megaprojects

• CoS socialized cost and risk (4 plants: $40B excess)
• Competitive generators more prudent, incur their own mistakes

2010s: coal retrofits vs. gas builds

• Market decisions outperform CoS

2020s: load growth, high entry barriers, government interference, CoS renewables start

“wholesale competition worked as intended…easy entry and exit markets of ERCOT, California, and the Northeast, and much more slowly in the vertically integrated Midwest, West and Southeast.” – DOE Electric Markets & Reliability Report lead (2017)

Figure (S&P Global): Rising PJM capacity prices lift returns for all assets. Map of PJM operating capacity by technology type (combined cycle, solar, wind turbine), with the Marcellus shale region highlighted.

Sources: Joskow (1997), Hogan (2001), Silverstein (2017), Yozwiak (2025), Davis & Weber (2025), Piper et al. (2024)

Operating Supply

• Secondary restructuring thesis, well examined
• Operating costs decline under competition
• Substantial nuclear operating efficiencies and innovation
• Trade gains: cost savings and reliability
• Lower environmental and RPS compliance costs
• Lower fossil fuel costs

Figure: Panel A, out-of-merit costs: observed production versus the local merit order, with the excess cost shaded. Panel B, gains from trade: local and foreign merit orders, showing import surplus, trade costs, and export surplus from interregional trade.

Sources: Davis & Wolfram (2012), Bushnell et al. (2017), Jha (2021), Cicala (2022), Cicala (2015), Chan et al. (2017)

Side Effects Unclear

Transmission and Distribution (T&D)

• Theoretical joint transmission-generation planning advantages of vertical integration
  • In practice: utility resource planning is rarely “integrated” usefully
• Vertical integration discourages efficient inter/regional transmission development
• Stronger T&D overcapitalization incentive for restructured utilities
• Transmission having stronger merchant ties

Interconnection

• Vertical integration has less coordination friction between utility-owned supply and delivery
• Utilities gatekeep third parties
• RTO interconnection problems often vestiges of vertical integration
• ERCOT (most restructured) has lowest interconnection barriers

Merits: competitive supply cost and risk benefits outweigh side effects of unbundling.

Sources: Hausman (2025), Hartman & Garza (2022), Peskoe (2021), Rand et al. (2026)

Price Effects

Competition lowers the cost curve, but needs dynamic market power mitigation to translate into downward price pressure.

• Energy and ancillary service markets = competitive results
• Capacity markets = mixed results (still outperform CoS procurement)
• Inflated average costs (utility) vs. marginal costs (market)

Exogenous factors outweigh industry structure (e.g., Southeast vs. Northeast).

• Exogenous costs: gas, labor, land, taxes, mandates, environmental compliance, surcharges, permitting, siting
• Intraregional: e.g., Midwest, Mid-Atlantic
• Large C&I vs. small residential differences

Chart: PJM Base Residual Auction (market) clearing price vs. Fixed Resource Requirement (utility) rate, 2020 to 2026, $/MW-day (approximate values read from chart). FRR rate stayed roughly $430 to $500 throughout: about $480 (2020), $465 (2021), $500 (2022), $450 (2023), $465 (2024), $450 (2025), $430 (2026). BRA clearing price: about $77 (2020), $140 (2021), $50 (2022), $34 (2023), $29 (2024), $270 (2025), $329 (2026). The BRA price remained below the FRR rate in every year shown.

Note: 2027 BRA clearing price: $333.44/MW-day. Market price cap effects on new entry bear watching.

Sources: Borenstein & Bushnell (2015), Hartman (2024), Kavulla (2025), Manzagol (2026)

Innovation

Chart: Research and Development (R&D) intensity by industry, 2023 (percent): Utilities 0.1; Manufacturing 5.5; Chemicals 10.6; Electrical equipment 4.3; Other 10.0.

CoS: CapEx-additive innovation, e.g., conventional nuclear, coal gasification.

Markets: cost-cutting or service enhancement innovation, e.g., operating efficiencies, combined cycle gas, solar plus storage.

Source: NASEM 2026 using NCSES 2025 data

Environment

Restructured markets induce emissions reductions by:

• Accelerating capital stock turnover
• Improving power plant efficiency
• Superior intraregional dispatch, interregional trade
• Lowering unconventional resource barriers (renewables, storage)
• Improving demand flexibility
• Responding more robustly to consumer environmental preferences
• Responding more robustly to environmental policy
• Innovation

Countervailing effects (limited):

• Nuclear retirements
• Isolated coal dispatch increase

Sources: Bushnell et al. (2017), Sharabaroff et al. (2009), Hartman (2017), Hartman & Brown (2026), Cicala (2025), Kiesling et al. (2026)

Customer Experience

Self-supply: markets lower barriers, align incentives (e.g., residential batteries, data center colocation).

Business customers: markets advantageous

• Larger users (benefits outweigh transaction costs)
• Products tailored to customer preferences
• Products tailored to customer configurations
• Off-grid, on-grid/hybrid, off-to-on grid (bridge)
• Far greater demand flex/response participation
• Commercial and industrials pushed for markets
• Data centers see load growth market advantage

Residential customers: mixed evidence

• Experience reflects implementation quality
• Quality leads to savings and customer preference matching
• Market failures: e.g., information deficiencies
• Government failure to “quarantine the monopoly” (e.g., monopoly-competitive market cross-subsidies)

Ohio case:

“electricity prices have increased following retail restructuring in all service territories in Ohio, with significant favourable welfare effects observed only in the Cincinnati area, where key policy implementation stages were not circumvented.” – Dormady et al. (2018)

2025: Ohio passes HB 15, eliminating cross-subsidies.

Sources: Giberson & Kiesling (2017), Lien (2008), Hibbard (2023), Kiesling & Macey (2026)

Retail Competition Quality Variance

Retail competition grades by state:

• A-: Texas
• B+: District of Columbia, Illinois, Ohio, Pennsylvania
• B: Delaware, Maine, Rhode Island
• B-: Massachusetts, New Hampshire, New Jersey
• C+: California, Connecticut, New York, Virginia
• C: Colorado, Maryland, Michigan
• C-: Arizona, Hawaii, Montana, North Carolina, Oregon, Vermont
• D+: Arkansas, Kansas, Kentucky, Louisiana, Nevada, South Carolina
• D: Alaska, Florida, Georgia, Indiana, Iowa, Missouri, New Mexico, North Dakota, Oklahoma, South Dakota, Tennessee, Washington, West Virginia
• D-: Idaho, Minnesota, Mississippi, Utah, Wisconsin, Wyoming
• F: Alabama

Sources: Villarreal et al. (2025)

Governance

Electoral governance

• CoS regulation: perverse political incentives
• Elevated cronyism, corruption cases
• Markets better, not immune to rent-seeking

Grid governance

• CoS regulation information asymmetry growing
• Transparency: market institutions outperform CoS
• Market institutions lead to complex market and government failure
• RTO accountability and incumbent-influenced rules
• Wholesale-retail disconnects (e.g., PJM)
• Greater state/federal market interventions

“rent-seeking strategies and the political maladies” – Alfred Kahn

Sources: Rubin (2011), Hartman & Haugh (2020), RTOGov initiative publications

Current Developments

Texas (ERCOT) best restructured, making it a performance outlier: unmatched innovative entry, demand flexibility, and load growth.

Rising marginal costs lead to market rate pressures. Huge investment and demand growth risk gives markets an advantage.

Large load responsiveness:

• Large load tariff efficiency vs. cost allocation
• Interconnection: ERCOT efficient, other RTO/ISOs delayed but improving
• CoS utilities self-selectively quick, scale limits (e.g., Entergy)
  • Gatekeep consumer-third party supplier deals
• Financing constraints (CoS utilities lack balance sheet expansion)
  • Fitch downgraded utilities to “deteriorating”

Emerging: deeper restructuring for business customers

• Benefits large loads, best insulates other customers (e.g., OK, VA, IN, WV, LA)
• Ratepayer Protection Pledge compliant: “build your own generation” (and transmission?)
• Residential remainder: restructuring vs. monopsony (i.e., generation competition, universal default service)

Chart: Electricity capacity additions, October 2025 to September 2026, net summer capacity in GW (completed plus expected; natural gas, solar, wind, oil, batteries, other): ERCOT 25.6; WECC 12.6; MISO 10.8; CAISO 7.0; SERC 5.0; PJM 4.8; SPP 4.2; Rest of US 3.7; ISO-NE 1.1; NYISO 0.4.

“One of the key differentiation factors for Entergy has been the up front cash received from customers that has allowed the company to only require ~15% equity financing of capex.” – Jefferies, 9/8/26

Sources: Fitch Ratings (2026), Electricity Customer Alliance (2026), Oklahoma Corporation Commission case 2025-75, FERC (2026), Campbell et al. (2026)

Takeaways

• Restructuring outperforms CoS when implemented well or moderately well.
  • High-quality CoS regulation may outperform low-quality restructuring.
• Will political economy support adequate restructuring implementation?
• Electricity economics increasingly market-conducive.
  • Generation diversity, customer partial self-supply configurations, storage utilization, retail in the digital age, merchant transmission, “privatize” resource adequacy (differentiated reliability).
• Would vertical integration under incentive-based regulation close much of the CoS-restructuring performance gap?
  • Incentive regulation hard to implement durably.

Contact

Devin Hartman. Email: dhartman@lhenergy.org. Website: lighthouseenergyinstitute.org