This casebook is not about whether nuclear baseload has a role in a diversified, decarbonising grid. That question has competent advocates on multiple sides and is outside this casebook's read. The decision this casebook prices is the recurring commit-to-completion authorisation the Vogtle project owners' boards faced from approximately 2009 onward, against a continuously-published independent monitoring record that was tracking cost and schedule in dollar terms every quarter for the entire duration of construction. The casebook treats the fifteen-year build as a single conversion arc — from the ~US$14 billion originally certified by the Georgia PSC in 2009 to the ~US$36.8 billion at completion in 2024 — and observes the structural feature that distinguishes Vogtle in the conversion library: the public-record cost data was not just available retrospectively. It was being published, by the project's own independent construction monitor reporting to the regulator, on a continuous quarterly cycle throughout the entire arc.

What happened, plainly

Between the 2009 Georgia PSC certification and the 2024 commissioning of Unit 4, Vogtle 3 & 4 ran fifteen years of construction across two presidential administrations, two CEO generations at the lead utility, the Chapter 11 bankruptcy of the original engineering-procurement-construction contractor, a US$3.68 billion parental-guarantee settlement from the contractor's parent company, and approximately thirty quarterly cycles of independent construction-monitoring reports filed with the Georgia Public Service Commission. The chronology below is built from Georgia Public Service Commission certification records, the project owners' SEC filings (Southern Company, Oglethorpe Power, MEAG Power), Westinghouse's bankruptcy docket (Southern District of New York, March 29, 2017), Toshiba's own corporate press releases on the parental-guarantee settlement, World Nuclear News, Power Magazine's published retrospective by the project's lead independent construction monitor, and contemporaneous reporting from Georgia and U.S. business press. Specific source attributions are in the Sources section.

When	Event	What was put on the public record
2008	Construction order placed with Westinghouse	Vogtle owners (Georgia Power, Oglethorpe Power, MEAG, Dalton Utilities) ordered two AP1000 reactors from Westinghouse Electric Company, with Toshiba as Westinghouse's parent providing a US$3.68 billion parental guarantee. Initial schedule projected Unit 3 in commercial operation by April 2016 and Unit 4 by April 2017.
Mar 2009	Georgia PSC certifies total project cost at ~US$14B	The Georgia Public Service Commission certified the Vogtle Units 3 & 4 expansion. Georgia Power's 45.7% share certified at approximately US$4.4 billion. Total project cost approximately US$14 billion. The first AP1000 reactor build in the U.S. and the first new U.S. nuclear construction project in over 30 years.
2009-2024	Independent monitoring reports filed quarterly	The Georgia Public Service Commission's independent construction monitor — a credentialed nuclear-construction professional reporting directly to the Commission — filed Vogtle Construction Monitoring (VCM) reports on a quarterly cycle for the duration of construction. Each report documented then-current cost estimates, schedule progress, and engineering and construction risks on the public regulatory record. The monitoring infrastructure was unbroken across the fifteen-year arc.
Mar 2013	Unit 3 construction officially begins	Site preparation had been underway since 2009; formal nuclear-construction commencement on Unit 3 dated March 12, 2013. Already approximately two years behind the original construction-start schedule, with cost-and-schedule slippage already visible in the monitoring record.
Nov 2013	Unit 4 construction officially begins	Formal nuclear-construction commencement on Unit 4 dated November 19, 2013. The two-unit construction was now both underway and both behind their originally-projected schedules.
2014-2016	Quarterly monitoring documents escalating cost and slipping schedule	Successive VCM reports documented engineering rework, AP1000 first-of-a-kind issues, contractor performance issues, and recurring schedule slippage. The dollar magnitude of the slippage was on the public record at each quarterly cycle. The commit-to-completion decision recurred at every cost-recovery filing and every capital-authorisation cycle for the project owners' boards.
2017 (Aug)	Georgia PSC and Georgia Power revise "reasonable cost" to ~US$7.3B	Following continued cost-and-schedule slippage and the Westinghouse bankruptcy, Georgia regulators and Georgia Power agreed that Georgia Power's reasonable share would be approximately US$7.3 billion — significantly above the original ~US$4.4 billion certified share. Implied total project cost now well above the original ~US$14 billion. The commit-to-completion decision was authorised at the revised figure.
Mar 29, 2017	Westinghouse files Chapter 11 in SDNY	Westinghouse Electric Company filed for Chapter 11 bankruptcy protection in the United States Bankruptcy Court for the Southern District of New York, citing approximately US$9 billion in losses on its U.S. nuclear-construction projects (Vogtle and V.C. Summer). The Vogtle owners lost their fixed-price-and-schedule contract protections. Construction continued under owner-led management, with Southern Nuclear taking over project management and Bechtel managing construction.
Jun 2017	Toshiba agrees parental-guarantee settlement of US$3.68B	Westinghouse's Japanese parent Toshiba reached agreement with the Vogtle owners on a US$3.68 billion parental-guarantee settlement, originally structured for installment payments from October 2017 to January 2021. The settlement partially offset the owner exposure created by the EPC contract loss, but did not stop the cost-escalation trajectory.
Jul-Aug 2017	V.C. Summer cancelled; Vogtle continues	The South Carolina V.C. Summer project — the only other U.S. AP1000 construction project, started at the same time as Vogtle — was cancelled by its owners (SCE&G and Santee Cooper) in July-August 2017 after the Westinghouse bankruptcy. Vogtle owners elected to continue. The decision to continue was made against a contemporaneous comparable that had just elected to cancel.
Dec 14, 2017	Toshiba pre-pays full US$3.68B parental guarantee	Toshiba completed early payment of the full US$3.68 billion parental-guarantee obligation to the Vogtle owners, well ahead of the original 2021 installment schedule. The funds eliminated some of the owners' immediate financial exposure but the cost-overrun trajectory continued on the construction record.
Aug 2018	Brookfield acquires Westinghouse from Chapter 11	Brookfield Business Partners completed its US$4.6 billion acquisition of Westinghouse from Toshiba, marking Westinghouse's exit from Chapter 11. Construction at Vogtle continued under the owner-led structure, with the cost-and-schedule trajectory still being documented quarterly in the VCM record.
2018-2022	Cost estimate continues climbing through subsequent VCM cycles	Successive VCM reports continued documenting cost escalation through the post-bankruptcy construction period. The commit-to-completion decision was authorised at each subsequent regulatory and board cycle against a documented escalation that the public monitoring record was tracking in dollar terms in real time.
Jul 31, 2023	Unit 3 enters commercial service	Vogtle Unit 3 entered commercial service approximately seven years behind the original April 2016 target. The first new U.S. nuclear reactor in over thirty years was now generating electricity.
Dec 2023	Georgia PSC approves further cost-recovery agreement	The Georgia Public Service Commission approved a financial agreement under which Georgia Power would cover at least US$2.6 billion of an expected US$10 billion in remaining construction and capital costs. Ratepayer cost-recovery framework continued to evolve as the total cost crystallised.
Apr 29, 2024	Unit 4 enters commercial service — conversion completes	Vogtle Unit 4 entered commercial service approximately seven years behind the original April 2017 target. Final total project cost approximately US$36.8 billion. The two-unit AP1000 expansion was complete. The most expensive power project in U.S. history.

"At completion, however, both the final cost of $36 billion and 15-year schedule were more than double original estimates."

Power Magazine — published retrospective by the project's lead independent construction monitor, April 2025

Why this matters to any board

The Vogtle case is in the conversion library because the underlying pattern is the cleanest possible version of a structural mechanism that recurs across every megaproject: a long-duration capital commitment, an independent monitoring infrastructure that tracks cost and schedule in real time in dollar terms, and a recurring commit-to-completion decision the project owners' boards face at every authorisation cycle. The same pattern travels through major dam and hydropower projects, transmission-line builds, refinery upgrades, mine developments, large defence procurements, regulated transit infrastructure, hospital and university capital projects, large public-private partnership infrastructure, and any number of asset classes where the question is not whether the original project was wrong but whether the next commit-to-completion authorisation is the right next decision against the documented record.

Three specific features make Vogtle structurally cleaner than the typical megaproject conversion case, and worth a board's attention precisely for that reason:

One — continuous independent monitoring in dollar terms. Most megaprojects do not have a regulator-appointed independent construction monitor publishing quarterly cost-and-schedule reports for the entire duration of construction. Vogtle did. The Georgia PSC's independent construction monitor filed VCM reports continuously from approximately 2009 through 2024. The cost-and-schedule data was not just available; it was on the regulatory record on a quarterly cycle for fifteen years.

Two — a contemporaneous decision-comparable that elected to cancel. The V.C. Summer project in South Carolina was the only other AP1000 construction project in the U.S., started at the same time, on the same reactor design, with the same EPC contractor and the same parent-company guarantee structure. When Westinghouse filed Chapter 11 in March 2017, V.C. Summer's owners elected to cancel by July-August 2017. Vogtle's owners elected to continue. The commit-to-completion decision at Vogtle from mid-2017 onward was being made against a contemporaneous, directly-comparable, near-identical project that had just made the opposite call.

Three — partial offset that did not change the conversion trajectory. Toshiba's US$3.68 billion parental-guarantee settlement, fully prepaid by December 2017, partially offset the owners' exposure but did not change the underlying cost-escalation curve. The conversion completed anyway, at approximately US$36.8 billion. The structural observation here is that financial offsets to specific contractor failures do not, by themselves, re-price a commit-to-completion decision; the underlying execution-environment exposure continues to compound on its own trajectory unless explicitly priced.

What Vogtle adds to the Resilience & Risk Management top-binding cluster, alongside Boeing (certification-vs-shipment irreversibility), Peloton (capacity-commitment), PG&E (maintenance-vs-distribution), and Vattenfall Moorburg (continuation-vs-exit on a hardened external trajectory), is the megaproject continuation structural variant: an independent monitoring infrastructure produces the data in real time; the commit-to-completion decision recurs at every authorisation cycle; and the absence of a single board-level priced read of the cumulative exposure means the conversion completes by default rather than by deliberate authorisation.

What the commit-to-completion decision has cost

The figures below are drawn from the Georgia Public Service Commission's certification and cost-recovery record, the project owners' SEC filings (Southern Company 10-K, Oglethorpe Power 8-K filings), Toshiba's own corporate press releases on the parental-guarantee settlement, Power Magazine's published retrospective, and contemporaneous reporting from Inside Climate News and Georgia Recorder. None of them is a NAVETRA figure. The cost section is a record of what has already moved on regulated-utility filings, regulatory records, and ratepayer accounts, not a forecast.

Total project cost escalation

~US$14B → ~US$36.8B

Originally certified total project cost at the March 2009 Georgia PSC certification was approximately US$14 billion across all four owners. Final total project cost at the April 2024 completion of Unit 4 was approximately US$36.8 billion. Approximately 2.6× the original estimate. The most expensive power project in U.S. history.

Schedule slip vs original in-service dates

~7 years

Original Unit 3 in-service target: April 2016. Actual: July 31, 2023. Original Unit 4 in-service target: April 2017. Actual: April 29, 2024. Approximately seven years behind on both units, with construction duration totalling fifteen years from the 2009 PSC certification.

Georgia Power certified-cost-share revisions

$4.4B → $7.3B → $10B+

Georgia Power's certified cost share moved from approximately US$4.4 billion at original March 2009 certification, to approximately US$7.3 billion under the 2017 "reasonable cost" agreement, with Georgia Power committed in December 2023 to cover at least US$2.6 billion of an expected US$10 billion in remaining construction and capital costs. The recurring cost-share revisions track the recurring commit-to-completion authorisation cycles.

Ratepayer cost-recovery impact

~$1,000 + ~$15/mo

Pre-construction surcharge collected from Georgia Power residential customers from 2009 to 2024 totalled more than US$1,000 per household for some customers. Post-completion base-rate increases approximately US$15 per month for a typical residential customer, recovered over decades. The cost of the commit-to-completion arc is being carried, in part, on ratepayer accounts across the asset's operating life.

Three additional structural costs are worth naming. First, the Toshiba parental-guarantee settlement of US$3.68 billion was a one-time partial offset that did not change the cost-escalation trajectory; the project still completed at ~US$36.8 billion total after the guarantee was paid in full. Second, the V.C. Summer cancellation crystallised a contemporaneous comparable outcome (the only other AP1000 project in the U.S., on the same timeline, with the same contractor) that elected to stop, providing a real-time decision benchmark at the same moment the Vogtle owners were authorising continuation. Third, the asset's underlying utility is real: approximately 2,200 MW of dispatchable low-carbon baseload generation now in service, with an expected 60-80 year operating life on the AP1000 design. The structural question the casebook prices is not whether the asset has value (it does), but whether the recurring commit-to-completion decision was authorised against a priced read of the cumulative exposure as the monitoring record documented it, or against the schedule and budget originally certified in 2009.

What the data showed, before each commit cycle authorised

The data that would have priced each commit-to-completion authorisation against the alternative paths (pause, restructure, cancel and recover salvage value) was on the public regulatory record at every quarterly monitoring cycle. None of the six data points below required private information from the Vogtle owners to read. All were in Georgia Public Service Commission filings, in SEC filings by the publicly-listed project owners, in Westinghouse's bankruptcy docket, in Toshiba's corporate press releases, or in major-press contemporaneous reporting.

Data point already in public record	What it described about the commit-to-completion decision
Quarterly VCM reports to the Georgia PSC, 2009-2024	The Georgia PSC's independent construction monitor was publishing cost-and-schedule data on a continuous quarterly cycle for the entire fifteen-year construction arc. Approximately thirty quarterly cycles. Each cycle was an opportunity to re-price the commit-to-completion authorisation against the then-current documented exposure.
Construction-start delays vs original schedule, 2013	Original construction start was implicit in the 2008 order; formal Unit 3 construction-start in March 2013 was already approximately four years after the order date, with Unit 4 in November 2013. Schedule slippage was structurally embedded before the AP1000 reactors were even being built.
Westinghouse Chapter 11 docket, March 29, 2017	The Southern District of New York bankruptcy filing detailed approximately US$9 billion in losses on Westinghouse's two U.S. AP1000 projects. The contractor's commercial viability on the AP1000 design was now an open court record. The loss of fixed-price and fixed-schedule contract protections was a structural shift in the Vogtle owners' risk position.
V.C. Summer cancellation, July-August 2017	The only other contemporaneous AP1000 construction project in the U.S. elected to cancel within four months of the Westinghouse bankruptcy. A direct, comparable, real-time decision benchmark on the same reactor design, the same contractor relationship, and the same parental-guarantee structure.
Georgia PSC 2017 "reasonable cost" revision to ~US$7.3B	The regulator's own formal upward revision of Georgia Power's reasonable cost share from ~US$4.4B to ~US$7.3B in August 2017 documented, on the regulatory record, that the originally-certified cost was no longer the cost being authorised. The commit-to-completion decision from that point forward was authorised at the revised figure, with the slippage trajectory still in motion.
Toshiba US$3.68B parental-guarantee settlement, June-December 2017	The full US$3.68 billion guarantee was paid by Toshiba by December 14, 2017, well ahead of the original installment schedule. The settlement value was on the public corporate record at Toshiba and on the project owners' SEC filings; the structural implication that the settlement partially offset but did not close the exposure gap was visible from the difference between the settlement value and the documented cost escalation.

None of these data points required private project access to read. The casebook makes no claim about whether they were assembled into a single board-level priced read at each commit-to-completion authorisation cycle inside the Vogtle owner organisations; that question cannot be answered from the public record alone. It observes that the dataset existed, was published continuously on a quarterly cycle in dollar terms by a regulator-appointed independent monitor for the entire duration of construction, and was available to any board authorising a continuation cycle on Vogtle from 2009 onward.

What NAVETRA does, briefly

NAVETRA produces one board-grade Operating Profit at Risk range on the execution environment a forward capital decision is landing into: one actuarially weighted, sector-validated figure, drawn from a corpus of 14,000+ assessments. The figure does not exist anywhere else in a standard governance stack. It prices the environment before the next authorisation cycle commits, on leading-indicator data the company already collects. It is patent-pending.

The artifact — the four-domain shape on a megaproject commit-to-completion case

The four domains below describe what NAVETRA's deliverable typically engages with on a commit-to-completion decision in this category: a long-duration megaproject with continuous independent monitoring infrastructure and recurring commit-to-completion authorisations. The reads are illustrative — no Operating Profit at Risk figure is assigned to the Vogtle owners, no domains are scored against them, and NAVETRA was not engaged by any of them. The shape is what a board sees on one page before the next authorisation cycle commits, not a finding about these companies.

Top binding · illustrative

Resilience & Risk Management

For a megaproject with recurring commit-to-completion authorisations against a continuously-published monitoring record, the largest exposure is the cumulative compounding of cost-and-schedule slippage across the authorisation cycles. Priced at each commit cycle against the documented quarterly slippage and against the alternative paths (pause, restructure, partial cancellation, full cancellation with salvage), the commit-to-completion decision becomes a deliberate authorisation rather than a default continuation. Vogtle is unusual in the conversion library specifically because the independent monitoring infrastructure produced cost data in dollar terms continuously throughout the entire fifteen-year arc.

Second binding · illustrative

Cross-Functional Collaboration

When the regulator's independent construction monitor, the EPC contractor's progress reporting, the owner organisations' internal capital-allocation processes, and the regulatory cost-recovery filings are all producing different views of the same project at the same time, the question for the board is not which view is right — it is whether they are reconciled into one priced read of the cumulative commit-to-completion exposure before the next authorisation cycle. The 2017 Westinghouse bankruptcy is a structurally clean instance: the contractor's commercial viability was now a bankruptcy-court fact, the V.C. Summer comparable was now a cancellation decision, and the parental-guarantee was now a specific dollar figure. Reconciling those three external facts into one priced read at that specific cycle, rather than at the next regulatory filing, would have been a different kind of board decision.

Third binding · illustrative

Knowledge Retention Sharing & Transfer

A first-of-a-kind reactor design (AP1000) being built for the first time in the U.S. carries irreducible engineering and construction risk that no execution-environment read can remove. What it can do is ensure that the lessons being recorded in the quarterly monitoring cycles, the engineering rework documentation, and the contractor performance reviews are reaching the next commit-to-completion authorisation as a priced exposure, not as engineering-status reporting. The structural insight from the parallel V.C. Summer cancellation is that the same engineering data, in the hands of a different ownership consortium, produced a different commit-to-completion decision — which suggests the data was strong enough to drive a board outcome, when it reached the board in the right form.

Fourth binding · illustrative

Executive Alignment

A multi-owner consortium with four separate corporate boards — a publicly-listed investor-owned utility (Georgia Power, Southern Company), a generation-and-transmission cooperative (Oglethorpe Power), a public power authority (MEAG), and a municipal utility (Dalton Utilities) — facing the same continuous commit-to-completion decision creates an alignment problem with structural features that an execution-environment read can describe. The four owner organisations have different governance structures, different stakeholder accountability mechanisms, different financial covenants, and different decision rights. A board-level priced read on the same project, at the same cycle, in the same dollar terms, lets each owner board authorise the next commit cycle deliberately rather than by consortium momentum.

The other six client-facing domains (Organization Alignment, Leadership Bandwidth, Team Effectiveness, Technology & AI Readiness, Talent & Hiring Alignment, Sales Readiness / Revenue Conversion) would be read alongside these four. The top binding pattern shown above is the one that travels with megaproject commit-to-completion decisions in this category, not with any individual project's situation.

What this casebook does not claim

Scope, regulatory adjudication, and exogenous-vs-endogenous split

Not characterising regulatory adjudication. The Vogtle 3 & 4 project has been the subject of extensive Georgia Public Service Commission cost-recovery proceedings, multiple PSC orders on the certified-cost revisions, and ongoing ratepayer-impact regulatory processes. The casebook does not characterise the PSC's prudence determinations, does not assert the cost-recovery framework was or was not appropriate, and does not re-adjudicate any specific regulatory proceeding. The recurring "reasonable cost" revisions are referenced as documented PSC orders, not as the casebook's analytical claim.

Not pricing the original sanction. The 2008 contractor selection, the 2009 Georgia PSC certification, and the AP1000 first-of-a-kind technology choice belong to a different policy era, a different state of the nuclear-construction industry, a different state of the U.S. baseload-generation mix, and predate the Westinghouse bankruptcy and the V.C. Summer cancellation. The casebook does not characterise those original decisions and does not assert the original sanction was wrong on the information available at the time. The casebook prices the recurring commit-to-completion decisions, framed as a single conversion arc from the ~US$14 billion 2009 certification to the ~US$36.8 billion 2024 completion.

No NAVETRA engagement, no figure assigned. NAVETRA was not engaged by Southern Company, Georgia Power, Oglethorpe Power, MEAG Power, Dalton Utilities, Southern Nuclear, the Georgia Public Service Commission, or any party to the project. No Operating Profit at Risk figure is assigned. No client-facing domains are scored against any of the project owners. The illustrative four-domain shape in the artifact section is a description of what NAVETRA's deliverable looks like for a megaproject commit-to-completion decision of this category, not a finding about any project owner.

Exogenous vs endogenous. AP1000 first-of-a-kind engineering risk, the Westinghouse contractor failure, the broader U.S. nuclear-construction industry capability gap (the first new U.S. nuclear in over thirty years means the construction workforce, supply chain, and regulatory inspection capacity were rebuilding in real time), and the nuclear-sector inflation context were exogenous. Their precise magnitudes and timings were not forecastable in real time with high precision. The casebook does not claim that any specific exogenous shock would have been timed by a priced read. The endogenous question is whether the recurring commit-to-completion authorisation was reconciled against the continuously-published monitoring record at each cycle. That is a board-seat question, not a forecasting question.

Attribution. The body text references the Georgia Public Service Commission's independent construction monitor by role only ("the Commission's independent construction monitor," "the Commission's lead independent construction monitor"). The pull-quote is sourced from a published retrospective in Power Magazine authored by the project's lead independent construction monitor in April 2025; the magazine's publication is the public-record reference, and the attribution in body text is to the role rather than to the individual. Other roles — predecessor utility CEOs, the Georgia PSC commissioners, the Westinghouse and Toshiba executives, the bankruptcy court personnel — are referenced by role or institution only. The casebook does not characterise the conduct of any individual; it references roles, institutional positions, and documented public statements as recorded in regulatory filings, SEC filings, bankruptcy-court records, and major-press reporting.

Forward, not retrospective scoring. The casebook does not characterise what any project owner's board did, did not do, or now must do. It describes a commit-to-completion decision the owners' boards recurringly faced, references the public-record monitoring data that was on the table at each cycle, and observes the conversion path that followed. The structural argument — that a continuous independent monitoring record was being published quarterly in dollar terms throughout the entire fifteen-year arc — is verifiable from primary regulatory and SEC sources; the inference about how that monitoring record was or was not engaged inside any specific owner organisation at any specific cycle is not something a third party can answer from filings alone.

Price the next commit cycle before the conversion completes by default.

For a CEO, board, or board risk committee sitting with a recurring commit-to-completion authorisation on a long-duration megaproject with continuous monitoring infrastructure and a documented cost-and-schedule trajectory, NAVETRA produces one board-grade Operating Profit at Risk range on the execution environment that decision is landing into. Actuarially weighted, sector-validated, drawn from a corpus of 14,000+ assessments. The figure does not exist anywhere else in the standard governance stack, and it is needed before the next commit cycle authorises.

Run the free NAVETRA™ Risk Scan

The Risk Scan is free and takes minutes. To discuss a specific decision directly, contact admin@purplewins.io or mjohl@purplewins.io.

NAVETRA^™

The path to the true north of a company

NAVETRA produces the board-grade Operating Profit at Risk range a board can read before a forward capital commitment hardens — one actuarially weighted, sector-validated figure, drawn from a corpus of 14,000+ assessments. The figure does not exist anywhere else in the buyer's stack.

Quantify

Quantify execution risk, see the hotspots, build the priority plan.

Invest

Human-capital investments as quantified business cases against the execution environment.

Align

AI deployment aligned from an outcome lens, so pilot dollars fund sustainable AI.

Learn

Personalised learning for every individual, rooted in execution realities.

Ten client-facing domains Organization Alignment · Executive Alignment · Leadership Bandwidth · Team Effectiveness · Knowledge Retention Sharing & Transfer · Technology & AI Readiness · Cross-Functional Collaboration · Talent & Hiring Alignment · Sales Readiness / Revenue Conversion · Resilience & Risk Management

Request a NAVETRA™ Risk Scan →

Sources & References

This casebook is built from the Georgia Public Service Commission's certification and cost-recovery records, the project owners' SEC filings (Southern Company, Oglethorpe Power 8-K filings), Westinghouse's Chapter 11 bankruptcy docket (Southern District of New York), Toshiba's own corporate press releases on the parental-guarantee settlement, published retrospective material from the project's lead independent construction monitor in Power Magazine, and contemporaneous reporting from World Nuclear News, Inside Climate News, Georgia Recorder, and Georgia Public Broadcasting.

Primary Regulatory and SEC Sources

Georgia Public Service Commission — Vogtle 3 & 4 certification and cost-recovery record, 2009-2024. Primary source for the March 2009 certification at approximately US$14 billion total project cost, the August 2017 "reasonable cost" revision to approximately US$7.3 billion for Georgia Power's share, the December 2023 financial agreement under which Georgia Power covers at least US$2.6 billion of expected US$10 billion in remaining construction and capital costs, and the continuous quarterly Vogtle Construction Monitoring (VCM) reports filed throughout the fifteen-year construction arc.
Georgia Public Service Commission — Vogtle Construction Monitoring docket
Oglethorpe Power Corporation — Form 8-K filing, October 2017, "Vogtle 3 & 4 Status Update." SEC primary source for the Westinghouse bankruptcy impact ("Westinghouse, the project's EPC contractor declared bankruptcy on March 29. As a result, owners lost fixed price and schedule protections set-forth in our contract"), the US$3.7 billion total investment status through June 30, 2017, the US$3.68 billion Toshiba parental-guarantee settlement terms, the owner-consortium percentages (Georgia Power 45.7%, Oglethorpe Power 30%, MEAG 22.7%, Dalton Utilities 1.6%), and Southern Nuclear's takeover of project management.
sec.gov — Oglethorpe Power Corporation Form 8-K, FY2017
Southern Company and Georgia Power SEC filings, 2009-2024. 10-K and 10-Q filings disclosing the project's cost-estimate progression, the Westinghouse-bankruptcy impact disclosures, and the certified-cost-share revisions over the construction period.
SEC EDGAR — Southern Company and Georgia Power annual and quarterly reports
Westinghouse Electric Company LLC Chapter 11 bankruptcy filing, March 29, 2017, United States Bankruptcy Court for the Southern District of New York. Primary source for the bankruptcy filing date, the approximately US$9 billion in losses on U.S. AP1000 projects (Vogtle and V.C. Summer), and the structural shift in the Vogtle owners' contractual position.
Westinghouse Electric Company LLC Chapter 11 docket, SDNY, March 2017

Toshiba Parental-Guarantee Primary Sources

Toshiba Corporation press release, December 14, 2017, "Toshiba Completes Full and Early Payment of Guarantee Obligations for Vogtle Nuclear Construction Project." Primary corporate-record source for the US$3.68 billion total parental-guarantee obligation, the June 2017 settlement agreement, the original October 2017 to January 2021 installment schedule, the December 14, 2017 full early payment to Georgia Power as agent for the owners, and the structural framework of the parental-guarantee settlement.
global.toshiba/ww/news/corporate/2017/12/pr1401.html
World Nuclear News, multiple articles, 2017-2018. Source for the Westinghouse Chapter 11 timeline, the Brookfield Business Partners acquisition of Westinghouse for approximately US$4.6 billion in August 2018, the parental-guarantee installment-schedule mechanics, and the V.C. Summer cancellation in July-August 2017 for comparative context.
world-nuclear-news.org — Vogtle, Westinghouse, and Toshiba coverage 2017-2018

Independent Monitor Retrospective and Outcome Reporting

Power Magazine — "What Was Learned from Building New Nuclear Reactors?", published retrospective by the project's lead independent construction monitor reporting to the Georgia Public Service Commission, April 2025. Source for the published retrospective framing ("At completion, however, both the final cost of $36 billion and 15-year schedule were more than double original estimates"), the Georgia Power 45.7% share of approximately 1,020 MW of new capacity, the approximately 2,200 MW combined Units 3 & 4 output, and the independent-monitor role context.
powermag.com/what-was-learned-from-building-new-nuclear-reactors
Inside Climate News — "Two Years After Completion, Plant Vogtle Still Looms Over the Nuclear Debate," May 2026. Source for the US$36.8 billion final cost framing, the seven-years-behind-schedule outcome, the "most expensive power project in U.S. history" characterisation, the approximately US$1,000 cumulative pre-construction surcharge per Georgia Power household 2009-2024, and the approximately US$15/month base-rate increase for typical residential customers post-completion.
insideclimatenews.org/news/10052026/plant-vogtle-nuclear-debate-two-years-after-completion
Georgia Recorder — "Biden administration, Georgia officials applaud debut of Plant Vogtle expansion," May 31, 2024. Source for the Unit 3 and Unit 4 commissioning timing, the fifteen-year construction duration, the Georgia Public Service Commission cost-share progression (~US$4.4 billion original → ~US$7.3 billion 2017 reasonable cost → at least US$2.6 billion of expected US$10 billion December 2023 agreement), and the regulatory-framework context.
georgiarecorder.com/2024/05/31/biden-administration-georgia-officials-applaud-debut-of-plant-vogtle-expansion
Georgia Public Broadcasting — "A second new nuclear reactor is completed in Georgia," April 29, 2024. Source for the Unit 4 commercial-service date, the four-owner-consortium operational handoff, and the corporate-statement context around the project's strategic positioning by the lead utility.
gpb.org/news/2024/04/29/second-new-nuclear-reactor-completed-in-georgia-the-carbon-free-power-comes-at-high
Nuclear Costs Track Record reference materials. Source for the originally-projected Unit 3 (April 1, 2016) and Unit 4 (April 1, 2017) commercial-service target dates, the actual Unit 3 (July 31, 2023) and Unit 4 (April 30, 2024) commercial-service dates, and the comparison framework for the AP1000 cost escalation.
nuclearcosts.org/track-record
Wikipedia — "Vogtle Electric Generating Plant" reference article. Source for the operator structure (Southern Nuclear), the formal construction-start dates (Unit 3 March 12, 2013; Unit 4 November 19, 2013), the AP1000 reactor specifications (~2,200 MW combined nameplate), the cooling-tower configuration, and the comparative reference to the 1976-1989 Vogtle Units 1 & 2 construction cost escalation ($660 million estimated to $8.87 billion actual).
en.wikipedia.org/wiki/Vogtle_Electric_Generating_Plant

Important Notice & Disclaimer

This casebook has been prepared by Purple Wins for informational and thought-leadership purposes only. It does not constitute financial, investment, legal, regulatory, environmental, or engineering advice, and should not be relied upon as the basis for any investment, business, governance, regulatory, or operational decision without independent professional verification.

This is a conversion casebook in the NAVETRA™ library. NAVETRA™ was not engaged by Southern Company, Georgia Power, Oglethorpe Power Corporation, MEAG Power, the City of Dalton, Dalton Utilities, Southern Nuclear, the Georgia Public Service Commission, Westinghouse Electric Company, Toshiba Corporation, Brookfield Business Partners, or any party to the Vogtle 3 & 4 project. The casebook does not claim access to any non-public information from any of these parties. No Operating Profit at Risk figure is assigned. No client-facing domains are scored against any party. The casebook makes no claim that NAVETRA™ would have surfaced or detected any matter. The illustrative four-domain shape in the artifact section describes the typical binding pattern for a megaproject commit-to-completion decision of this category, not a finding about any specific project owner.

Not characterising regulatory adjudication

The Vogtle 3 & 4 project has been the subject of extensive, ongoing Georgia Public Service Commission cost-recovery proceedings, including multiple PSC orders on certified-cost revisions, prudence reviews, and ratepayer-impact regulatory processes. The casebook does not characterise the PSC's prudence determinations, does not assert that the cost-recovery framework is or is not appropriate, and does not re-adjudicate any specific regulatory proceeding. Specific characterisations advanced by parties in those proceedings are not relied on as the casebook's analytical claims. The casebook's load-bearing structural observation — that an independent construction-monitoring record was being published quarterly in dollar terms throughout the construction arc — is verifiable from the Georgia PSC's own docket records and is the only structural claim being made.

Not pricing the original sanction

The 2008 contractor selection, the 2009 Georgia Public Service Commission certification, and the AP1000 first-of-a-kind technology choice belong to a different policy era, a different state of the nuclear-construction industry, and predate both the Westinghouse Chapter 11 bankruptcy (March 29, 2017) and the V.C. Summer cancellation (July-August 2017). The casebook does not characterise those original decisions and does not assert that the original sanction was wrong on the information available at the time. The casebook prices only the recurring commit-to-completion decisions the owners' boards faced from approximately 2010 onward, framed analytically as a single conversion arc from the ~US$14 billion 2009 certified total project cost to the ~US$36.8 billion 2024 completion cost.

No allegation of wrongdoing

This casebook does not allege wrongdoing, misconduct, breach of duty, breach of fiduciary obligation, regulatory violation, or any other violation of law by Southern Company, Georgia Power Company, Oglethorpe Power Corporation, MEAG Power, the City of Dalton or Dalton Utilities, Southern Nuclear Operating Company, the Georgia Public Service Commission or its commissioners, Westinghouse Electric Company LLC, Toshiba Corporation, Brookfield Business Partners, the project's independent construction monitor, or any individual associated with any of these organisations. Specific characterisations attributed in this casebook to documented regulatory orders, SEC filings, bankruptcy-court records, corporate press releases, and major-press reporting are referenced as documented public materials; this casebook does not extend those characterisations beyond their published scope. References to the Georgia Public Service Commission's "reasonable cost" revisions are characterised as PSC orders on the regulatory record, not as findings of imprudence or fault.

Methodological scope

The casebook addresses the endogenous, governable part of the situation: the recurring commit-to-completion authorisation on a long-duration megaproject whose cost-and-schedule trajectory was being independently documented in dollar terms on a continuous quarterly cycle. Broader exogenous factors — AP1000 first-of-a-kind engineering risk, the U.S. nuclear-construction industry capability gap (the first new U.S. nuclear in over thirty years implies a workforce, supply-chain, and regulatory-inspection capacity rebuilding in real time), the Westinghouse contractor failure, the nuclear-sector inflation context, and the broader U.S. baseload-generation policy environment — are described as context only and are explicitly outside the casebook's read. The casebook does not predict the post-completion operating performance of Vogtle Units 3 & 4, does not characterise the operational safety or reliability of the AP1000 design, and does not characterise the future regulatory or ratepayer-recovery framework. The asset's underlying value as approximately 2,200 MW of dispatchable low-carbon baseload generation with a 60-80 year expected operating life is referenced as a publicly-documented characterisation, not as a casebook recommendation or endorsement of nuclear baseload generation as an investment category.

General

All financial figures and corporate-decision characterisations are drawn from publicly available regulatory filings (Georgia Public Service Commission), SEC filings (Southern Company 10-K and 10-Q, Oglethorpe Power 8-K and 10-K filings), bankruptcy-court records (Westinghouse Electric Company LLC Chapter 11 filing, Southern District of New York, March 29, 2017), corporate press releases (Toshiba Corporation, December 14, 2017), published retrospective material in industry publications (Power Magazine), and reputable contemporaneous reporting. Currency figures denominated in U.S. dollars are nominal as reported; Toshiba's parental-guarantee figure of approximately US$3.68 billion is also recorded in Toshiba's primary corporate communications in Japanese yen terms (approximately ¥412.3 billion). Cost-share percentages and figures attributed to the four-owner consortium are drawn from documented SEC filings and PSC records. Purple Wins has made reasonable efforts to represent those sources accurately but accepts no liability for inaccuracies, omissions, or misinterpretations arising from reliance on this casebook.

NAVETRA™ is a product of JTS Inc. (Jawaahar Talent Solutions Inc., Ontario), operated under the Purple Wins brand. Purple Wins is not affiliated with, endorsed by, or acting on behalf of Southern Company, Georgia Power Company, Oglethorpe Power Corporation, the Municipal Electric Authority of Georgia (MEAG Power), the City of Dalton or Dalton Utilities, Southern Nuclear Operating Company, the Georgia Public Service Commission, Westinghouse Electric Company LLC, Toshiba Corporation, Brookfield Business Partners, the U.S. Department of Energy, the Nuclear Regulatory Commission, or any other party referenced in this casebook. All trademarks remain the property of their respective owners. © Purple Wins. NAVETRA™ is a trademark of JTS Inc. Patent-pending.

Vogtle Units 3 & 4.The capital risk on the commit-to-completion decision that the Georgia Public Service Commission's own independent construction-monitoring record had been pricing every quarter for fifteen years.