GH-G-03: Ghana's Power Sector — Dumsor and the IPP Debt Trap (1966–2026)

Status: DRAFTWords: 9,152

1. Key Takeaways

  • Ghana's power sector is the country's longest-running unfinished governance project and, since the 2010s, the largest single hole in its public finances. The sector's history runs in a discernible spiral: a hydro-dependent foundation laid by Nkrumah's Akosombo Dam (commissioned 22 January 1966, one month before his overthrow); recurring drought-and-supply crises — 1983–84, 1997–98, 2006–07, the great dumsor of 2012–16, and a renewed round in 2024–25 [TBD-VERIFY: severity and duration of the 2024–25 load-shedding episode]; emergency procurement in each crisis on terms that outlive the crisis; and an accumulating chain of arrears — distributor to generators, generators to fuel suppliers, everyone ultimately to the sovereign — that by the mid-2020s constituted the largest quasi-fiscal exposure on Ghana's balance sheet and a standing pillar of the IMF programme's structural conditionality (see GH-D-04, GH-D-02). The pattern this document explains is how a country moved from not enough power to too much contracted power without ever passing through enough — and why both states were fiscally ruinous.

  • The Akosombo foundation embedded three structural facts that have governed everything since. First, hydro-dependence: a single dam on a single river supplied the overwhelming majority of national generation for four decades, making the entire economy hostage to Volta-basin hydrology — a climate vulnerability built in at the founding and demonstrated in every drought since 1982. Second, the VALCO bargain: the dam was financed around a 1962 Master Agreement that committed roughly half its firm output to the Kaiser-led Volta Aluminium Company smelter at Tema at one of the lowest fixed electricity tariffs in the world [TBD-VERIFY: the original VALCO rate, commonly cited at c. 2.625 mills/kWh, and its 30-year lock-in terms], a deal whose renegotiation saga (1980s arbitration pressure, the 1985 rate revision, the 2003 shutdown and 2004 state takeover) is the original Ghanaian lesson in long-dated power contracts signed under duress. Third, the cheap-power illusion: Akosombo's near-zero marginal cost set political expectations of inexpensive electricity that no thermal-era cost structure could meet, founding the tariff politics that has under-priced power ever since.

  • Dumsor — the Akan compound for "off-and-on" — entered the national vocabulary during the 2012–16 crisis, but it names a recurrence, not an event. Each of the five major crises followed the same anatomy: a hydrological or fuel-supply shock hit a system with no reserve margin and a financially hollow distribution segment; rationing and load-shedding followed for one to four years; the government of the day procured emergency thermal capacity at premium prices; and the under-pricing and under-collection that had left the system fragile were left substantially unreformed. The 2012–16 episode was the longest and most damaging: a hydro shortfall combined with gas-supply failure — the West African Gas Pipeline's chronic Nigerian shortfalls and its August 2012 severing by a ship's anchor off Lomé, plus delays in monetising Jubilee-field associated gas — produced roughly four years of scheduled and unscheduled outages whose economic toll was estimated at on the order of 2% of GDP a year [TBD-VERIFY: ISSER and World Bank estimates of dumsor's 2014–15 cost, commonly cited in the US$320m–680m/yr and 1.8–2% of GDP range].

  • The political bill for the 2012–16 crisis was paid by John Mahama, who acquired the tag "Mr Dumsor", staked his presidency on a promised fix ("I will fix it" became the contested slogan of 2015–16), and in December 2016 became the first incumbent president defeated in the Fourth Republic — a defeat to which dumsor, alongside the broader economic squeeze, was central by every contemporary account (see GH-H-PRES-05; his 2024 return is GH-E-01). The crisis also produced a distinctive social history: the generator class divide (firms and households that could afford diesel self-generation versus those that could not), the dumsor-vendor economy of generator importers, fuel hawkers, and phone-charging kiosks, and a celebrity-led protest politics (the May 2015 #DumsorMustStop vigil fronted by actress Yvonne Nelson) that marked the issue's escalation from inconvenience to national political fact.

  • The crisis response of 2014–16 created the next crisis. Under emergency conditions Ghana signed a wave of independent power producer (IPP) and emergency-power contracts — the Dubai-brokered AMERI 250MW fast-track deal (US$510m over five years, later found to carry an estimated US$150m intermediary markup over the Metka subcontract price [TBD-VERIFY: the AMERI contract value, the markup estimate, and the 2018 renegotiation collapse that cost the energy minister his job]), the Turkish Karpowership floating plants, AKSA, Cenpower, Amandi, Early Power and others — overwhelmingly on take-or-pay terms indexed to the dollar, obliging the state to pay for contracted capacity whether or not it was consumed. When supply normalised after 2017, Ghana found itself with roughly 2,300MW of contracted capacity above peak demand and annual payments for unconsumed power that the Akufo-Addo government put at around US$500 million a year, plus a parallel obligation for unused gas [TBD-VERIFY: the 2019 Mid-Year Budget's excess-capacity figures and the ~2,300MW excess-capacity estimate]. The emergency-procurement trap — crisis terms locked in for ten to twenty years — is the sector's defining policy failure and a textbook case in the African IPP literature.

  • Beneath the contracting layer sits the distribution-segment haemorrhage. The Electricity Company of Ghana (ECG) loses on the order of a quarter to a third of the power it buys to technical, commercial, and collection losses [TBD-VERIFY: ECG aggregate ATC&C loss rates, variously reported at 25–32% across the 2015–2025 period], and what it collects has been chronically insufficient because tariffs are a political price: every Fourth Republic government has at some point suppressed tariffs into an election, refilling the arrears pool. The flagship attempt to fix ECG — the 2019 Power Distribution Services (PDS) concession under the US Millennium Challenge Corporation's Compact II — collapsed within months when the demand guarantees securing the handover were found to be invalid [TBD-VERIFY: the PDS suspension (July 2019) and termination (October 2019) sequence and the forfeiture of c. US$190m in remaining MCC compact funds], ending the boldest privatisation experiment in the sector's history and chilling the private-participation debate for half a decade.

  • The arrears chain — ECG → IPPs and VRA → fuel suppliers → banks and the sovereign — accumulated into a sector debt overhang on the order of US$2–3 billion by the mid-2020s [TBD-VERIFY: energy-sector arrears stock estimates, which vary by date and definition; figures of US$1.5–3bn are cited across 2019–2025 official and IMF sources], serviced through the Energy Sector Levies Act (ESLA, 2015) fuel levies and the ESLA PLC bond programme (roughly GH₵10bn issued from 2017 [TBD-VERIFY]), managed through the Energy Sector Recovery Programme (2019) and the Cash Waterfall Mechanism allocating ECG's collections among creditors, and finally embedded as structural conditionality in the 2023 IMF Extended Credit Facility — quarterly PURC tariff adjustments, a moratorium on new unsolicited PPAs, and renegotiation of the worst contracts (see GH-D-04). The 2023–26 renegotiations with the major IPPs produced standstills and restructured terms whose final shape remains in motion [TBD-VERIFY: outcomes of the 2023 IPP standstill agreements and the Mahama-era renegotiations of Karpowership, AKSA, and Cenpower terms].

  • The sector's institutional architecture is a reform pioneer's cautionary tale. Ghana unbundled early and textbook-style — VRA generation (1961), GRIDCo transmission (carved out 2006–08), ECG/NEDCo distribution, a Public Utilities Regulatory Commission and Energy Commission (both 1997) — and attracted one of sub-Saharan Africa's earliest and largest IPP fleets. Yet unbundling without cost-reflective pricing and collection discipline simply distributed the deficit along the chain: each interface (PURC tariff decisions, ECG payments to VRA and IPPs, VRA payments for gas) became a point where political under-pricing was converted into somebody's receivable. The gas layer added a second take-or-pay tier — the ENI/Vitol Sankofa offshore gas project, backstopped by World Bank guarantees, obliges Ghana to pay for roughly 90% of contracted gas regardless of offtake [TBD-VERIFY: Sankofa take-or-pay percentage and annual exposure] — creating the double take-or-pay problem: paying simultaneously for unused capacity and unused fuel.

  • Against the financial wreckage stands a genuine delivery achievement: the National Electrification Scheme (from 1989) raised access from roughly a quarter of the population to over 85% [TBD-VERIFY: access rate, commonly cited at 85–89% by 2024], among the highest in sub-Saharan Africa — and Ghana exports power to Togo, Benin, and Burkina Faso through the West African Power Pool even while its own sector drowns in arrears. The ledger is therefore genuinely two-sided: near-universal access and regional-exporter status delivered by the same arrangements that produced unpayable debts. The unresolved 2030s questions — demand growth meeting under-investment, the renewables stall (solar held back by the excess-capacity overhang), the Bui hydro-solar hybrid, and the nuclear ambition [TBD-VERIFY: status of Ghana's SMR/large-reactor vendor selections] — all run through the same unreformed fiscal core (see GH-O-01 §7, which sketches the cycle this document treats in full).

2. The Akosombo Foundation (1961–1990s)

2.1 The Dam as Nkrumah's Signature

The Volta River Project was the largest single act of the Nkrumah developmental state and the physical centrepiece of the industrialisation vision documented in GH-A-02. The idea predated Nkrumah — colonial-era surveys of a Volta dam and aluminium scheme go back to Albert Kitson's 1915 bauxite identification and the 1949–52 Anglo-colonial Volta River Scheme studies — but it was Nkrumah who carried it to financial close and construction, treating it explicitly as the energy foundation on which a manufacturing economy would be built. The Volta River Development Act (Act 46) of 1961 created the Volta River Authority (VRA) as an integrated developer-operator with responsibilities running from generation through lake transport, fisheries, and the resettlement of the displaced; the dam at Akosombo was built by the Italian contractor Impregilo between 1961 and 1965 at a cost on the order of US$130 million [TBD-VERIFY: total project cost figures range c. US$130m–$258m depending on whether the smelter, township, and transmission are included], financed by the World Bank, the United States, and the United Kingdom alongside Ghana's own contribution. Nkrumah commissioned the dam on 22 January 1966; the coup that ended his government came on 24 February, thirty-three days later. The conjunction is more than ironic: the dam was simultaneously the regime's most durable legacy and a monument to the cocoa-financed fiscal overreach that helped end it (see GH-G-02 on the cocoa squeeze that part-financed the era's ambitions).

The project's social cost was carried by the Volta basin's inhabitants. Lake Volta, at roughly 8,500 km² the largest man-made lake in the world by surface area, submerged some 700 villages and displaced approximately 80,000 people [TBD-VERIFY: resettlement figures of 78,000–84,000 across c. 700–740 villages are standard], whose resettlement in 52 new townships became one of the era's signature planned-development exercises and one of its quiet failures — inadequate farmland, broken livelihoods, and an unresolved compensation politics that persists in the Volta and Oti regions. The scholarly literature, from David Hart's The Volta River Project (1980) to Stephan Miescher's A Dam for Africa (2022), treats Akosombo as the paradigm case of high-modernist African development: genuinely transformative, genuinely extractive of its local losers, and built around a bargain with foreign capital whose terms shaped the sector for decades.

2.2 The VALCO Bargain

That bargain was the Volta Aluminium Company (VALCO) smelter at Tema. The dam's financing logic required an anchor customer large enough to guarantee revenue, and the 1962 Master Agreement with a consortium led by Kaiser Aluminum (90%) and Reynolds Metals (10%) supplied it: VALCO would build a smelter taking a large block of Akosombo's firm output — in the early decades commonly cited as 45–60% of generation [TBD-VERIFY: VALCO's offtake share of Akosombo output in the late 1960s–1970s] — at a fixed tariff widely cited at around 2.625 mills (US$0.002625) per kWh, among the lowest aluminium-smelting power prices in the world, locked for thirty years with limited escalation [TBD-VERIFY: the precise original rate, escalation clauses, and duration; the 30-year term with renewal options is standard in the literature]. The deeper asymmetry was structural: VALCO smelted imported alumina — Jamaican and elsewhere — rather than Ghanaian bauxite, so the integrated bauxite-alumina-aluminium chain that justified the scheme in Nkrumah's prospectus never materialised. Ghana supplied cheap power to a tolling operation whose upstream and downstream value accrued abroad; the refinery that would convert Ghanaian bauxite into alumina remains unbuilt sixty years later (the GIADEC saga, GH-O-01 §7).

The renegotiation saga ran for two decades. By the late 1970s the fixed rate had been rendered derisory by inflation and the oil-era repricing of energy; the PNDC government opened renegotiation from 1982, with the 1982–84 drought — which forced VALCO's pot lines down and demonstrated that the "surplus" power underpinning the 1962 deal no longer existed — supplying leverage. The 1985 revised agreement raised the rate several-fold [TBD-VERIFY: the 1985 renegotiated rate, commonly cited at c. 17 mills/kWh, against the original 2.625 mills] and introduced periodic review. The endgame came in the next drought era: VALCO curtailed and then ceased smelting in 2003 amid power shortages and price disputes, Kaiser sold its 90% to the Government of Ghana in 2004, and Alcoa (successor to the Reynolds share) exited in 2008, leaving the state owning an ageing smelter that has since operated intermittently at a fraction of capacity. The VALCO story is the sector's founding precedent on three points that recur throughout this document: long-dated contracts signed under financing duress outlive their circumstances; renegotiation is possible but takes decades and crises; and the industrialisation rationale for power infrastructure is the easiest promise to make and the hardest to deliver.

2.3 The Hydro Architecture and the VRA's Golden Age

For its first three decades the system was essentially one institution and one resource. The VRA generated (Akosombo's original 912MW across four units, expanded to six units and c. 1,020MW by 1972 and uprated to c. 1,038MW in the 2000s retrofit [TBD-VERIFY: capacity figures by phase]; the 160MW Kpong run-of-river plant downstream added in 1982), transmitted, and sold — to ECG for the south, to its own Northern Electricity Department (later NEDCo, created 1987) for the north, to VALCO, and to the Communauté Electrique du Bénin for Togo and Benin. Within this architecture the VRA built a reputation as one of Africa's best-run utilities: professionally managed, creditworthy, an engineering-career destination, and operationally autonomous in a way few Ghanaian state enterprises achieved. The National Electrification Scheme (1989) — a thirty-year plan to take the grid to every community above 500 inhabitants, accelerated by the Self-Help Electrification Programme — was executed substantially through this machinery and constitutes, with the access numbers cited in the Key Takeaways, the sector's great delivery success.

The golden age rested, however, on two wasting assets. The first was hydrological luck: a system drawing 99% of supply from one basin was a bet on rainfall, and the bet's downside was demonstrated in 1982–84 and periodically thereafter. The Volta's inflows are among the more variable of major African rivers, and the lake's storage — enormous in absolute terms — buffers perhaps two bad years, not three. Climate variability was thus built into the foundation, decades before anyone framed it as climate change; the September 2023 emergency spillage of Akosombo, which flooded downstream communities and displaced tens of thousands [TBD-VERIFY: the 2023 spillage displacement figures, commonly reported around 26,000–35,000], showed the same dependence operating in reverse. The second wasting asset was cheap-power politics: Akosombo's near-zero marginal cost allowed tariffs that bore no relation to the cost of the next megawatt. When demand growth exhausted the hydro surplus in the 1990s and thermal generation (fuel-cost-driven, dollar-denominated) had to be added, the gap between the political price of power and its economic cost opened — and it has never since closed.

3. The Dumsor Cycles (1982–2025)

3.1 The Recurrence Pattern

Ghana has experienced five major power crises: 1982–84, 1997–98, 2006–07, 2012–16, and 2024–25 [TBD-VERIFY: characterisation of 2024–25 as a major episode; it was materially shorter than 2012–16]. Each followed the same anatomy. A supply shock — drought in the first three, drought-plus-gas-failure in the fourth, fuel-finance-plus-gas in the fifth — hit a system running without reserve margin because under-pricing and under-collection had starved investment. Load-shedding followed, lasting one to four years. The government of the day procured emergency thermal capacity at crisis prices. And the structural causes were left substantially intact, because the politically rational response to a power crisis is to add supply visibly, not to raise tariffs and fix collection invisibly. The crises also distribute neatly across the political spectrum — PNDC (1982–84), NDC under Rawlings (1997–98), NPP under Kufuor (2006–07), NDC under Mahama (2012–16), NPP under Akufo-Addo with a tail into the NDC's return (2024–25) — which is analytically important: dumsor is not a party pathology but a system one.

1982–84 was the founding crisis, coinciding with the general economic catastrophe, famine, and the Nigerian expulsions documented in GH-B-03. The Sahelian drought drew the lake toward its minimum operating level; Akosombo's output was slashed, VALCO's pot lines shut, exports were curtailed, and rationing was imposed on an economy already in collapse. The crisis fed the ERP-era recognition that the hydro surplus was finite, and supplied the leverage for the 1985 VALCO renegotiation. 1997–98, under the elected Rawlings government, was the El Niño-driven sequel: the lake fell to record lows, months of scheduled outages followed, and the response set the thermal turn — completion of the first Takoradi Thermal Power Station units at Aboadze (the 330MW TAPCO plant, commissioned 1997–2000, followed by the TICO joint venture with CMS Energy, later TAQA) [TBD-VERIFY: commissioning dates and capacities of TAPCO/TICO phases], Ghana's first significant non-hydro generation and its first experience of dollar-denominated fuel costs flowing through a cedi-denominated tariff.

2006–07, under Kufuor, repeated the hydrology shock on a larger economy: roughly a year of load-shedding, VALCO shut again, and an emergency-procurement response that included the Tema thermal expansion, a mines-reserve plant, the long-delayed deployment saga of the Osagyefo power barge [TBD-VERIFY: the barge's procurement-to-deployment history], and two decisions of lasting consequence — the 2007 agreement with Sinohydro and China Exim Bank to build the 400MW Bui Dam (commissioned 2013), and acceleration of the West African Gas Pipeline as the future fuel solution. The 2006–07 crisis is also when the political template hardened: the opposition NDC campaigned on the government's energy failure in 2008, exactly as the NPP would on the NDC's in 2016.

3.2 The Great Dumsor, 2012–16: Anatomy

The 2012–16 crisis was the compound failure of both supply architectures at once. The hydro leg failed conventionally: poor inflows from 2012 drew the lake down toward minimum operating levels by 2014–15, cutting output from Akosombo, Kpong, and the newly commissioned Bui. The gas leg failed structurally. The West African Gas Pipeline, operational from 2008–11 after long delays, had never delivered its contracted Nigerian volumes reliably — upstream supply shortfalls, domestic Nigerian obligations, and payment disputes kept flows chronically below the c. 120 MMscf/d Ghana planned around [TBD-VERIFY: contracted versus delivered WAGP volumes] — and in August 2012 a ship's anchor severed the pipeline off Lomé, cutting flows entirely for roughly a year. Domestic gas was supposed to fill the gap: Jubilee field associated gas, processed at the Atuabo gas plant built by Sinopec under the China Development Bank facility. But Atuabo was not commissioned until 2015, years behind the schedule assumed when the thermal fleet was planned [TBD-VERIFY: the originally planned versus actual Atuabo commissioning dates], and in the interim the thermal plants burned light crude oil that the cash-strapped utilities — ECG's receivables crisis was by now chronic — could not reliably afford. The result was the worst of all configurations: hydro short, gas absent, liquid fuel unaffordable, and a distribution company too broke to pay for whichever fuel was available.

The outages ran, with varying intensity, from 2012 into 2016 — at the 2015 trough on schedules of 24 hours off, 12 on in parts of Accra [TBD-VERIFY: the published load-shedding timetables' ratios], with unscheduled outages on top. The economic toll estimates converge on a severe number: the Institute of Statistical, Social and Economic Research (ISSER) put 2014 losses around US$680 million, roughly 2% of GDP, and World Bank and CDD-Ghana work produced figures of similar magnitude for 2015, with production losses concentrated in manufacturing and agro-processing SMEs that could not afford self-generation [TBD-VERIFY: the ISSER/World Bank loss estimates and their methodologies; per-day figures of c. US$2.1m circulated contemporaneously]. Firm-level surveys recorded businesses losing 40%+ of production hours; the period's growth slowdown — GDP growth falling from the oil-boom peaks above 9% to around 2–4% by 2014–16 — had dumsor as a leading cause alongside the parallel fiscal crisis that took Ghana into the 2015 IMF programme.

3.3 The Social Texture and the Political Bill

Dumsor generated its own social economy. The generator divide stratified the country by capacity to self-supply: banks, hotels, and large firms ran industrial diesel sets (at two to four times the grid tariff); the middle class bought Chinese petrol generators whose hum became the night sound of Accra; the majority simply lost the hours. A dumsor-vendor economy flourished — generator and inverter importers, rechargeable-lamp and power-bank sellers, phone-charging kiosks, fuel hawkers serving queues — alongside the costs that did not show in GDP: spoiled cold-chain inventory, study hours lost, hospital wards on candles between generator cycles, and a fire-and-fume casualty toll from improvised power [TBD-VERIFY: documented generator-related fire/CO incidents during the crisis]. Culturally, dumsor became the lens for everything: radio phone-in staple, sermon metaphor, comedy genre, and — in the May 2015 #DumsorMustStop vigil led by actress Yvonne Nelson and other celebrities in Accra — the occasion for a new, social-media-organised, expressly non-partisan protest politics that prefigured the civic mobilisations of the following decade.

The political bill was paid in full. President Mahama, who had inherited the crisis's onset in 2012, made the fatal rhetorical moves himself: complaining publicly that Ghanaians had taken to calling him "Mr Dumsor", and promising in his 2015 State of the Nation address to fix the crisis rather than manage it — a formulation ("I will fix it") that opposition communicators converted into a measuring stick. The emergency procurement of 2015–16 (Section 4) did substantially end load-shedding by the 2016 election season, and the government campaigned on having fixed dumsor; but the electorate's experienced reality of 2013–15, compounded by the cedi crisis, the IMF return, and public-sector pay restraint, produced the December 2016 defeat — Akufo-Addo's 53.7% to Mahama's 44.4%, the largest margin of the Fourth Republic to that date and the first ejection of a sitting president (see GH-H-PRES-05; GH-D-01 covers the presidency in full). The lesson every subsequent government internalised was not "price power correctly" but "never be in power during load-shedding" — an incentive that explains both the 2014–16 procurement binge and the tariff suppressions that followed.

3.4 The 2024–25 Round

The recurrence came on schedule. From late 2023 into 2024, under the fiscally prostrate late-Akufo-Addo government (GH-D-07 context), intermittent load-shedding returned — driven not by hydrology but by the financial layer: arrears to fuel suppliers and IPPs interrupting gas and liquid-fuel procurement, transmission and distribution faults, and disputes in which generators (most visibly the Chinese-owned Sunon Asogli plant, owed hundreds of millions of dollars) shut down or threatened to over non-payment [TBD-VERIFY: the Sunon Asogli shutdown episodes and arrears figures, c. US$259m cited in 2024]. The government resisted publishing a load-shedding timetable — itself a politically charged act, since a timetable would concede that dumsor had returned by name — while ECG and GRIDCo traded blame; Yvonne Nelson marked the symmetry by protesting again. The episode carried into the Mahama government's first year, with the new energy minister warning publicly in 2025 of fuel-stock and financing fragility [TBD-VERIFY: the 2025 supply situation and whether scheduled load-shedding recurred under the new government]. The 2024–25 round matters analytically because it demonstrated the cycle's new form: the binding constraint is no longer generation capacity — Ghana has contracted excess — but the sector's cash flow. Dumsor's fifth round was a balance-sheet crisis wearing a power-cut costume, which is the bridge to Section 4.

4. The Emergency-Power Binge and the IPP Debt Trap (2014–2026)

4.1 The Crisis-Procurement Wave

A government facing nightly blackouts and an election does not run competitive tenders; it signs what can be delivered fastest. Between roughly 2014 and 2016 Ghana contracted a wave of emergency and fast-track generation whose collective terms define the sector's present. The emblematic deal was AMERI: a 2015 build-own-operate-transfer agreement with Africa & Middle East Resources Investment Group of Dubai for 250MW of trailer-mounted GE TM2500 aeroderivative turbines at Aboadze, at US$510 million over five years. Parliamentary and civil-society scrutiny (led by ACEP and IMANI) subsequently established that AMERI — a broker with no power-sector track record, associated with a Dubai royal — had subcontracted the entire works to the Greek contractor METKA for a sum around US$360 million, implying an intermediary margin in the region of US$150 million for, in effect, signing first [TBD-VERIFY: the AMERI and METKA contract values and the markup estimate; figures derive from the 2017–18 ministerial review and parliamentary debate]. The Akufo-Addo government's attempted remedy compounded the lesson: a 2018 renegotiation that would have novated the agreement to a Mytilineos vehicle on extended terms was withdrawn under scrutiny as worse than the original, and Energy Minister Boakye Agyarko was dismissed in August 2018 over its handling [TBD-VERIFY: the renegotiation's terms and the dismissal's stated grounds]. The plant ultimately transferred to the VRA at the end of its term [TBD-VERIFY: transfer completion date, reported c. 2024].

The second emblem was Karpowership: the Turkish floating-power operator whose powership docked at Tema in late 2015 under a ten-year PPA with ECG, initially around 225MW burning heavy fuel oil, later replaced by a larger vessel (c. 450–470MW) relocated in 2019 to the Sekondi naval base to run on domestic gas [TBD-VERIFY: vessel capacities, the relocation date, and the PPA's pricing terms, which have never been fully published]. Around these headline deals came the broader fleet: AKSA (Turkish, c. 370MW HFO), Cenpower Kpone (c. 350MW combined cycle, 2019), Amandi (c. 200MW, Bridge/Early Power and others following), atop the earlier Sunon Asogli plant (Shenzhen Energy/China-Africa Development Fund, 200MW from 2010, expanded to 560MW) — which collectively converted Ghana from a two-generator system into host of one of sub-Saharan Africa's largest IPP fleets [TBD-VERIFY: individual plant capacities and commissioning dates]. Two features were near-universal: dollar-denominated tariffs, transferring currency risk to a cedi-earning buyer, and take-or-pay capacity charges, obliging payment for contracted availability whether or not the power was dispatched. Under blackout conditions both terms were rational for investors and arguably unavoidable for the state; the trap lay in their duration — ten to twenty years of crisis pricing for what was a three-to-four-year crisis.

4.2 The Excess-Capacity Paradox

Demand growth slowed with the post-2014 economy precisely as the contracted fleet came online, and by 2018–19 Ghana had swung from deficit to structural surplus. The Akufo-Addo government's own framing of the inheritance, in the July 2019 Mid-Year Budget Review, became the canonical statement of the paradox: installed and contracted capacity far above peak demand — the excess commonly put around 2,300MW against a peak load then under 3,000MW [TBD-VERIFY: the excess-capacity estimate and the contemporaneous peak-demand figure] — with annual payments of approximately US$500 million for capacity not consumed, and a further c. US$250 million committed for gas not used [TBD-VERIFY: the Mid-Year Review's figures and subsequent Energy Sector Recovery Programme restatements], with projections that unreformed obligations would cost US$12.5 billion over 2020–23. The government announced conversion of take-or-pay contracts to take-and-pay and a moratorium on new PPAs; the conversions were partial and contested by investors holding enforceable contracts [TBD-VERIFY: which PPAs were actually converted or renegotiated 2019–22]. The paradox deserves precise statement: Ghana was now paying crisis prices for insurance against a crisis that the payments themselves were helping to cause, since capacity charges flowed into tariffs and arrears that kept the sector too broke to maintain and fuel the system reliably — the mechanism behind the 2024–25 round.

The same period produced the PDS fiasco, the failed attempt to fix the demand side. Under the US Millennium Challenge Corporation's second compact (signed 2014, c. US$498 million, the largest US compact in the power sector), ECG's operations were to pass to a private concessionaire for twenty years. Power Distribution Services Ghana — a consortium led by Manila Electric (Meralco) with Ghanaian partners holding the politically mandated 51% local share — took over ECG's southern operations on 1 March 2019. By July 2019 the government suspended the concession upon discovering that the demand guarantees securing ECG's assets, purportedly from Qatar's Al Koot Insurance, were invalid — issued without authority [TBD-VERIFY: the precise defect in the guarantees and the conflicting findings of the FTI Consulting review]; in October 2019 the concession was terminated outright. The US side disputed the termination's grounds and withheld the compact's remaining tranche (c. US$190 million) [TBD-VERIFY: the forfeited amount and MCC's stated position]. Whatever the underlying facts — the episode remains contested between accounts emphasising genuine fraud and accounts emphasising political discomfort with the concession — the institutional consequence was unambiguous: the most serious private-participation attempt in ECG's history collapsed inside eight months, the loss-reduction investment it was to carry evaporated, and "ECG privatisation" became politically radioactive for half a decade.

4.3 The Arrears Chain and the Sector Debt

The financial architecture of the trap is a chain of receivables. ECG buys power from VRA and the IPPs but collects insufficient revenue (Section 5); it therefore pays generators partially and late. The IPPs carry the arrears at penalty rates or threaten shutdown; VRA, paid worse than the IPPs (whose contracts have teeth), in turn underpays its fuel suppliers — the Ghana National Gas Company, the Nigerian gas shippers through WAGP, and liquid-fuel traders — and GNPC ends up carrying the Sankofa gas bill the state guaranteed. Each link's deficit is real money owed to entities with bank loans, so the arrears metastasise into the financial system; and because many obligations are sovereign-guaranteed or World Bank-backstopped, they consolidate, eventually, onto the state. Estimates of the stock at various dates run from c. US$1.5 billion in the 2019 ESRP baseline to the US$2–3 billion range cited through 2023–25 official and IMF documentation, with the sector's annual financing shortfall — the gap between cash collected and obligations falling due — running on the order of US$1.5–2 billion by 2025 [TBD-VERIFY: arrears-stock and annual-shortfall figures by year; definitions vary between legacy debt, current arrears, and projected shortfall].

The state's management instruments accumulated in layers. The Energy Sector Levies Act (2015) consolidated fuel levies into a dedicated debt-service stream; ESLA PLC (2017) securitised that stream through cedi bonds — c. GH₵6 billion in the first issues, expanding toward GH₵10 billion [TBD-VERIFY: cumulative ESLA issuance] — to pay down legacy energy and TOR debts, in effect refinancing the sector's past at the pump. The Energy Sector Recovery Programme (ESRP, 2019), designed with the World Bank, set the reform menu: cost-reflective tariffs, ECG loss reduction, least-cost procurement, renegotiation, and the Cash Waterfall Mechanism (2020) — a monthly, formula-based allocation of ECG's collections among generators and suppliers, designed to replace discretionary (and politically influenced) payment ordering with transparent pro-rata distribution. Then the 2022 fiscal collapse subsumed everything: under the 2023 IMF Extended Credit Facility (GH-D-04, GH-D-02), the energy sector's shortfall became a defined fiscal risk with structural benchmarks attached — maintenance of the PURC's quarterly tariff-adjustment mechanism, an updated ESRP, no new unsolicited generation contracts, and reduction of the shortfall on a published path [TBD-VERIFY: the specific energy-sector structural benchmarks across ECF reviews]. The 2023 standstill agreements with five major IPPs averted formal default and opened renegotiations — pursued with renewed intensity by the Mahama government from 2025 over the Karpowership, AKSA, and Cenpower terms — whose outcomes remain partially unresolved as of mid-2026 [TBD-VERIFY: renegotiation status; see GH-E-04 for the Year-One treatment].

5. The Institutional Architecture and Its Pathologies

5.1 The Unbundled Chain

On paper Ghana's sector is a textbook reformed African power market. Generation is plural: the VRA's hydro and thermal assets, the Bui Power Authority (the 2007 Act creating a second state hydro developer), and the IPP fleet. Transmission is a ring-fenced monopoly: the Ghana Grid Company (GRIDCo), carved out of the VRA by the 2005 power-sector reforms and operational from 2008, owns and operates the national interconnected transmission system. Distribution is two regional monopolies: ECG (established 1963 from the colonial Electricity Department, corporatised 1997) for the southern two-thirds where demand concentrates, and NEDCo (a VRA subsidiary) for the north. Regulation is dual: the Public Utilities Regulatory Commission (PURC, Act 538 of 1997) sets tariffs and service standards; the Energy Commission (Act 541 of 1997) licenses and plans. The unbundling — among the earliest in sub-Saharan Africa — was genuinely implemented, not merely legislated, and the entry of IPPs from the Takoradi joint ventures onward made Ghana a celebrated case in the reform literature of the 2000s.

The pathology is that unbundling distributed the deficit without eliminating it. Each interface in the chain became a valve where political under-pricing converted into somebody's receivable. The PURC-to-ECG interface: tariff awards have repeatedly lagged cost — the commission's major resets (the 2015–16 increases, the c. 27% and 30% adjustments of 2022, and the quarterly mechanism operated under IMF conditionality since 2022–23 [TBD-VERIFY: the quarterly mechanism's start date and adherence record]) have been followed by election-season pauses, and the cedi's depreciations between adjustments silently devalue every award against dollar-denominated costs. The ECG-to-everyone interface: ECG's aggregate technical, commercial, and collection losses — network losses, theft and meter tampering, unbilled consumption, and non-payment (with public institutions historically among the worst payers) — have run in the 25–32% range [TBD-VERIFY: ATC&C loss rates by year], meaning a quarter to a third of purchased power generates no revenue. The dispatch interface: with excess contracted capacity, which plants run (and therefore whose fuel is bought and whose capacity sits idle but paid) becomes a discretionary decision with large commercial consequences, a standing integrity risk. The Mahama government's 2025–26 response — accelerated prepaid and smart metering, enforcement against non-paying public bodies, the cash waterfall's tightening, and a revived but reframed ECG private-sector-participation process [TBD-VERIFY: the 2025–26 ECG PSP transaction's structure and status] — is the latest attempt to close the master valve; the record counsels respect for how often that attempt has failed.

5.2 The Gas Layer and the Double Take-or-Pay

Beneath the power chain sits a gas chain with its own contracting pathology. Domestic gas comes from three offshore sources — Jubilee and TEN associated gas processed at the Atuabo plant (Ghana National Gas Company, commissioned 2015), and the Sankofa-Gye Nyame (OCTP) non-associated gas field developed by ENI and Vitol with GNPC, onstream from 2018. Sankofa was bankable only with extraordinary support: World Bank guarantees of around US$700 million (IDA/IBRD) plus MIGA cover — the Bank's largest-ever guarantee package for an African energy project at signing [TBD-VERIFY: guarantee amounts and structure] — and a gas sales agreement obliging Ghana to pay for roughly 90% of contracted volumes whether taken or not [TBD-VERIFY: the take-or-pay percentage and annual dollar exposure]. In the early years Ghana's downstream could not absorb the contracted volumes — pipeline interconnection between the western enclave and the Tema demand centre lagged, and the excess-capacity power fleet needed less gas than planned — so the state paid for undelivered gas while simultaneously paying for undispatched capacity: the double take-or-pay problem, a fiscal pincer in which both the fuel and the plant are insured at the sovereign's expense against a demand that under-pricing has made unbankable. The WAGP import leg from Nigeria, the original gas hope, remains operational but chronically under-delivered and periodically disrupted by Nigerian supply and payment disputes; the net effect by the mid-2020s was a paradoxical gas position — contracted surplus, operational fragility — mirroring the power position exactly. Declining investment in the oil fields that supply associated gas (GH-O-01 §7) threatens the volumes precisely as power demand growth absorbs the excess.

5.3 The Renewables Stall

Ghana's renewables record is a study in how contract overhang blocks technological transition. The Renewable Energy Act (Act 832) of 2011 set a 10% renewables share target (excluding large hydro) for 2020, with feed-in tariffs and purchase obligations; the target was missed by an order of magnitude — non-hydro renewables remained around 1–2% of generation by the mid-2020s [TBD-VERIFY: the renewables share and the target's rebasing to 2030] — and the feed-in regime was effectively suspended. The proximate cause is rational: a system paying hundreds of millions of dollars a year for idle thermal capacity has no financial logic for adding intermittent capacity, however cheap solar has become; every new solar PPA would deepen the take-or-pay hole before it shallowed the fuel bill. The exceptions prove the diagnosis: utility solar has advanced mainly where it displaces fuel cost without new private contracting — the Bui Power Authority's hydro-solar hybrid, pairing a c. 250MW planned solar park (the first c. 50MW phases commissioned from 2020) with the dam's reservoir as a virtual battery [TBD-VERIFY: Bui solar installed versus planned capacity], the VRA's smaller solar plants at Navrongo and Kaleo, and distributed rooftop systems among firms hedging both tariffs and reliability. The structural insight matters for the 2030s: Ghana's energy transition is blocked less by finance or resource than by the unamortised sins of the emergency-procurement era — the excess-capacity overhang must be worked off, renegotiated, or grown into before large-scale renewables become contractually rational.

6. The Political Economy of Power

6.1 Electricity as the Swing-Voter Issue

Electricity is arguably the Fourth Republic's most reliable swing issue, because it is the rare governance output every voter experiences identically, immediately, and verifiably. The correlations are suggestive even where causation is layered: the 2008 election followed the 2006–07 crisis and produced the narrowest alternation in Ghanaian history (Mills by under 41,000 votes); the 2016 election followed the great dumsor and produced the Fourth Republic's first incumbent defeat; the 2024 election followed the cedi-and-debt collapse with the 2024 load-shedding round as accompaniment, and produced the largest opposition margin yet (GH-E-01). No serious analysis attributes these outcomes to power alone — each ran through a general economic crisis — but in each case electricity supplied the experiential proof of the macro story: the lights going out is the moment an abstract fiscal crisis becomes a household fact. The corollary shapes behaviour in office: governments treat visible supply as existential and invisible solvency as deferrable, which is the political root of every pathology in Sections 4 and 5.

Tariffs carry the same politics with the opposite sign. Ghana has a tariff-protest tradition running from the trade-union mobilisations against utility increases in the 1990s, through #OccupyFlagstaffHouse (2014) and the dumsor vigils — protests where tariff and reliability grievances fused — to the recurring confrontations between the PURC and organised labour, AGI industrialists (who pay among the region's highest industrial tariffs to cross-subsidise residential lifeline bands while suffering its least reliable supply during crises), and residential consumers over each adjustment round [TBD-VERIFY: the major tariff-protest episodes and the industrial-residential cross-subsidy structure]. The electoral cycle is visible in the tariff series: real-terms tariff declines or freezes cluster before elections — the 2018 residential cut, the pre-2016 and pre-2024 pauses [TBD-VERIFY: the specific election-adjacent tariff decisions] — and corrections cluster after them, with the 2022–23 IMF-anchored quarterly mechanism representing the first attempt to take the tariff out of the electoral calendar entirely. Whether quarterly adjustments survive the 2028 campaign season is, as GH-O-01 §7 notes, the single best indicator for the sector's trajectory.

6.2 The Procurement-Corruption Nexus

Crisis procurement is corruption's preferred weather, and the power sector has supplied the Fourth Republic's longest-running procurement controversies. The AMERI affair is the archetype — an intermediary's nine-figure margin on a sole-sourced emergency deal, exposed by civil-society scrutiny (ACEP, IMANI), litigated through parliamentary committees, and then nearly compounded by a renegotiation that would have extended the rent — and it bracketed both parties: signed under the NDC, mishandled in renegotiation under the NPP, with a minister dismissed. Around it sit a genre: the Karpowership PPA's never-fully-published terms; the early-2010s GYEEDA and SUBAH-era pattern of unexamined sole-sourcing extending into energy; the SML revenue-assurance contract controversy in petroleum measurement [TBD-VERIFY: the SML contract's scope and the 2024 KPMG audit findings]; and the PDS guarantees, whichever account of them one credits. Parliament's role has been real but asymmetric — PPAs require parliamentary ratification under Article 181, making the chamber a genuine veto point, but ratification battles track party lines, and the minority's forensic energy (John Jinapor against AMERI's renegotiation; NPP figures against PDS) reliably inverts when the benches swap. The deeper structural point belongs to the comparative literature: take-or-pay contracts signed in emergencies are rent-extraction's ideal instrument, because urgency forecloses competition, complexity hides margins, and the payment obligation arrives years after the signing government has banked the political credit for ending the blackouts.

6.3 The Regional Dimension and the Access Achievement

Ghana's power system has always been regional. Akosombo was designed with export commitments to Togo and Benin (through the CEB), and Ghana remains a structural exporter through the West African Power Pool (WAPP) interconnections — to CEB, to Burkina Faso (the Bolgatanga-Ouagadougou line), and onward — even at the depths of its own financial crisis. The paradox of exporting power while unable to pay for fuel is less irrational than it appears (exports earn hard currency against marginal cost, and the excess-capacity position makes export the best available use of idle plant) but it has generated recurrent domestic political friction, particularly during load-shedding episodes, and its receivables add a further link to the arrears chain when neighbours — historically CEB — fall behind on payment [TBD-VERIFY: export volumes and cross-border arrears positions]. WAPP integration is also the sector's most plausible long-run discipline: a functioning regional market would price Ghanaian excess capacity into Sahelian deficits, monetising the overhang — but the pool's payment-security and transmission constraints have kept that promise, like much else in the sector, ahead of its delivery.

The achievement side of the ledger deserves its full weight. From an access rate around 25% when the National Electrification Scheme launched in 1989, Ghana reached over 85% access by the mid-2020s [TBD-VERIFY: the access figure and series] — behind only a handful of sub-Saharan states, far ahead of Nigeria (c. 55–60%) and of the regional average — with urban access near-universal and the grid extended through the SHEP self-help mechanism to communities that mobilised their own poles and wiring. Electrification was delivered by both parties across every alternation, one of the Fourth Republic's genuinely bipartisan continuities, and its developmental and welfare effects (study hours, vaccine cold chains, phone charging as the platform for mobile money and the digital economy) are foundational to everything in GH-G-01's social-policy story. The honest framing is therefore not failure but unfunded success: Ghana connected its population faster than it built the revenue model to serve them, universalising a service priced below cost — the welfare-state pattern, visible equally in the NHIS and Free SHS (GH-G-01, GH-K-02), of delivery outrunning sustainability.

7. Comparative Perspective and the 2030s Questions

7.1 Ghana in the African Power-Reform Literature

Ghana occupies a specific niche in the comparative literature: the unbundling pioneer turned cautionary case. The 1990s–2000s reform orthodoxy — unbundle, regulate independently, invite IPPs, privatise distribution — was implemented in Ghana more faithfully than almost anywhere on the continent, and the canonical surveys (Eberhard, Gratwick, Morella and Antmann's Independent Power Projects in Sub-Saharan Africa, 2016, and the successor literature) long treated Ghana alongside Kenya as the IPP success story. The post-2015 Ghanaian record forced the literature's revision: it demonstrated that the orthodox sequence fails — indeed, fails expensively — when implemented atop a distribution segment that cannot collect and a tariff that politics will not allow to reflect cost. The comparison set sharpens the point. Nigeria privatised distribution wholesale in 2013 and got the same arrears chain with private faces, plus an access and supply record far worse than Ghana's. Kenya kept KPLC majority-state but listed, contracted IPPs more conservatively, and ran cost-reflective tariffs longer — yet by the 2020s faced its own IPP-cost political revolt and renegotiation commission. South Africa never unbundled, and Eskom's trajectory — load-shedding from 2007, a debt mountain absorbed onto the sovereign balance sheet, and a state-capture procurement record — delivered the continent's largest version of the same end-state by the opposite institutional route (see ZA-O-01 §3 for the Eskom treatment). The convergent lesson across all four: institutional form matters less than the political economy of the tariff and the collection rate. Ghana's distinctive contribution to the genre is the excess-capacity variant — proof that a country can contract its way from deficit straight to a differently shaped insolvency.

7.2 The Emergency-Procurement Trap as a Genre

The Ghanaian IPP record also reads as a chapter in a global genre: the emergency power contract. The pattern — blackout crisis, sole-sourced fast-track deals with dollar-indexed take-or-pay terms, post-crisis excess capacity, decade-long fiscal hangover, renegotiation under multilateral pressure — recurs from Pakistan's 1994 and 2015 IPP waves (renegotiated 2020–24 under IMF pressure on terms Ghanaian negotiators have studied) to Indonesia's post-1997 PLN contracts, Senegal and Tanzania's emergency-lease scandals (Richmond/IPTL), and Karpowership's client list across Africa and beyond. The genre's lessons are consistent and were all violated in Ghana's 2014–16 wave: cap emergency-contract tenor to the crisis horizon; separate the capacity decision from the crisis (standing procurement frameworks, competitively pre-qualified); never let the buyer of last resort be an entity (ECG) whose insolvency is the crisis's underlying cause; and publish the contracts, because secrecy is where the margin lives. Ghana's 2019 PPA moratorium, the standstill renegotiations, and the IMF-era ban on unsolicited proposals are the standard genre remedies; the genre's track record suggests they hold until the next crisis-election conjunction, which is precisely the discipline the 2028 cycle will test.

7.3 The 2030s Questions

Three questions frame the sector's next decade, all sketched in GH-O-01 §7 and given fuller statement here. First, demand growth meets the overhang: peak demand growing at 6–10% a year [TBD-VERIFY: demand-growth rates] will absorb the contracted excess by the late 2020s, converting today's surplus problem into tomorrow's deficit problem unless investment resumes — and the next megawatts must be financed by a sector whose creditworthiness the arrears chain has destroyed, in a post-default sovereign context with no fiscal cushion (GH-D-02). The danger case is exact: a pre-2028 tariff freeze refills the arrears pool, a gas or hydrology shock meets a hollow sector, and the sixth dumsor round triggers a third emergency-procurement wave on worse terms than the second. Second, the transition financing question: Ghana's least-cost expansion path is contested between domestic gas (cheap if upstream investment holds, stranded if it does not), solar-plus-storage (cheap and falling, but blocked by the overhang and the absence of a creditworthy offtaker), regional trade, and the nuclear ambition — Nuclear Power Ghana's IAEA-milestone programme, with announced vendor selections pairing a US small-modular-reactor consortium (NuScale/Regnum Technology Group) and a Chinese large-reactor track [TBD-VERIFY: the 2024–25 vendor-selection announcements, their contractual status, and financing — an MoU is not a project], whose 2030s delivery would repeat the Akosombo pattern of a single transformative bet if it proceeds and the long Ghanaian tradition of announced megaprojects if it does not. Third, the ECG question, the master variable: whether the revived private-sector-participation process, the metering programme, and the cash waterfall finally make the distribution segment solvent — because every other reform in the sector's sixty-year history has eventually drowned in ECG's receivables, and there is no version of the 2030s that works without fixing the till.

8. Conclusion

The Ghanaian power sector's sixty-year arc admits a compact statement. Nkrumah built a magnificent dam around a bad contract and a cheap-power expectation; the dam outlived him by generations and so did both flaws. The hydro architecture worked until demand and drought exhausted it, failing first in the general catastrophe of 1983 and then on a roughly decadal cycle thereafter. Each failure was answered with the politically rational response — visible emergency supply — rather than the economically necessary one — solvent pricing and collection — and the 2012–16 crisis, the longest and most consequential, converted that pattern into a fiscal trap: a fleet of dollar-indexed take-or-pay contracts that turned shortage into a paid-for surplus, layered atop a gas chain with the same disease, all draining through a distribution company that loses or fails to collect a third of what it sells. The arrears chain that resulted is now welded into the sovereign's IMF-supervised balance sheet, and the fifth dumsor round of 2024–25 demonstrated the end-state: a country with excess generating capacity suffering blackouts because nobody in the chain can pay for fuel.

Yet the record refuses a simple failure narrative, and the corpus's tone discipline requires the refusal be substantive. The same sector delivered near-universal electrification a generation ahead of its income peers, anchored an industrial zone at Tema, exports power to three neighbours, attracted one of Africa's deepest IPP markets, and built regulatory institutions (PURC, the Energy Commission, the parliamentary ratification check) that — however imperfectly — have made Ghanaian power contracting more scrutinised, and its scandals more exposed, than almost anywhere in the region. The sector's pathology is not state incapacity; it is a specific, identifiable political equilibrium in which reliability is existential for incumbents, tariffs are existential for voters, and the gap between them is financed by arrears that come due on someone else's watch. The dumsor cycle is that equilibrium's weather system.

Whether the cycle breaks is therefore a political question wearing technical clothes, and it has a date: the 2028 election will reveal whether the IMF-era quarterly tariff mechanism — the first institutional attempt to remove the price of power from the campaign calendar — survives contact with a competitive Fourth Republic campaign. If it survives, and the IPP renegotiations and ECG turnaround hold, Ghana would have done what no government since 1966 has managed: carried a solvent power sector through an alternation. If it does not, the sixth round is already written — the same anatomy, a worse balance sheet, and no borrowing cushion. Sixty years after Nkrumah stood at Akosombo and declared the energy foundation of an industrial Ghana laid, the dam still generates, the smelter still waits for its refinery, and the foundation still awaits the building — which is, compressed to a sentence, the whole history this document has told.

Primary Sources Consulted:

  1. Hart, David. The Volta River Project: A Case Study in Politics and Technology. Edinburgh University Press, 1980.
  2. Moxon, James. Volta: Man's Greatest Lake. André Deutsch, 1969.
  3. Miescher, Stephan F. A Dam for Africa: Akosombo Stories from Ghana. Indiana University Press, 2022.
  4. Republic of Ghana. Volta River Development Act, 1961 (Act 46); Renewable Energy Act, 2011 (Act 832); Energy Sector Levies Act, 2015 (Act 899).
  5. Eberhard, Anton, Katharine Gratwick, Elvira Morella, and Pedro Antmann. Independent Power Projects in Sub-Saharan Africa: Lessons from Five Key Countries. World Bank, 2016.
  6. Kumi, Ebenezer Nyarko. "The Electricity Situation in Ghana: Challenges and Opportunities." Center for Global Development Policy Paper 109, 2017.
  7. Institute of Statistical, Social and Economic Research (ISSER). The Economics of Electricity Power Outages in Ghana (dumsor cost assessment), University of Ghana, 2015. [TBD-VERIFY: exact title and publication year of the ISSER dumsor study]
  8. Ministry of Finance, Republic of Ghana. Mid-Year Fiscal Policy Review of the 2019 Budget Statement (excess-capacity and take-or-pay figures), July 2019.
  9. Ministry of Energy / World Bank. Energy Sector Recovery Program (ESRP), 2019, and subsequent updates.
  10. International Monetary Fund. Ghana — Requests for and Reviews under the Extended Credit Facility Arrangement, Country Reports, 2023–2025 (energy-sector annexes and structural benchmarks).
  11. Africa Centre for Energy Policy (ACEP). Analyses of the AMERI agreement, the PDS concession, and energy-sector debt, 2015–2025.
  12. IMANI Centre for Policy and Education. Briefings on AMERI, PDS, and power-sector procurement, 2015–2020.
  13. Public Utilities Regulatory Commission (PURC). Tariff decisions and quarterly adjustment publications, 1998–2026.
  14. Energy Commission of Ghana. National Energy Statistics and Energy Outlook for Ghana, annual editions.
  15. Volta River Authority. Annual Reports and the VRA's published Akosombo/Kpong operational histories.
  16. Millennium Challenge Corporation / Millennium Development Authority. Ghana Power Compact (Compact II) documentation and termination statements, 2014–2019.
  17. Parliament of Ghana. Hansard and committee proceedings on the AMERI agreement ratification and renegotiation, 2015–2018, and PPA ratifications under Article 181.
  18. World Bank. Sankofa Gas Project guarantee documentation (IDA/IBRD/MIGA), 2015, and Ghana energy-sector public expenditure reviews.
  19. Gratwick, Katharine N., and Anton Eberhard. "Demise of the Standard Model for Power Sector Reform and the Emergence of Hybrid Power Markets." Energy Policy 36 (2008).
  20. Ackah, Charles, et al. Studies on dumsor's firm-level and welfare impacts, ISSER/IGC Ghana, 2015–2017. [TBD-VERIFY: specific IGC working-paper citations]
  21. Reuters, Bloomberg, Africa Confidential, and Ghanaian press (Daily Graphic, Joy News, Citi FM) reporting on the PDS suspension/termination (2019), the IPP standstills (2023), Sunon Asogli shutdowns (2024), and the 2024–25 load-shedding round.
  22. Nuclear Power Ghana / Ministry of Energy statements on the nuclear power programme and vendor selections, 2023–2025.

Related Documents:

  • GH-A-02: The Nkrumah Era and the First Republic (1957–1966) — the Volta River Project's political context and the industrialisation vision
  • GH-G-01: Ghana's Social Policy — The NHIS, Free SHS, and the Welfare-State Experiment — the parallel delivery-versus-sustainability pattern
  • GH-G-02: Cocoa Political Economy — COCOBOD and the Farmer-State Bargain — the sibling policy-domain treatment; cocoa surplus as Volta-era development finance
  • GH-O-01: Ghana Megatrends — The 2030s Questions — §7 sketches the power-sector cycle this document treats in full
  • GH-D-02: The 2022 Domestic Debt Exchange Programme and the 2023 IMF Extended Credit Facility — the fiscal crisis that consolidated energy-sector debt onto the sovereign
  • GH-D-04: 2022 Domestic Debt Exchange and the IMF Programme — the programme's energy-sector structural conditionality
  • GH-E-01: The 7 December 2024 Election and the Mahama Return — the 2024 election context and the dumsor-era political memory
  • GH-E-04: Mahama Year One and the Mid-Term (2025–2026) — the IPP renegotiations, ECG PSP debate, and tariff politics in progress
  • GH-H-PRES-05: John Dramani Mahama — the "Mr Dumsor" tag, the 2016 defeat, and the 2024 return
  • GH-K-02: The 2017 Free SHS Decision and Its Fiscal Politics — the companion case of popular delivery outrunning financing
  • GH-C-02: back-reference added by symmetry sweep
  • GH-H-PRES-03: back-reference added by symmetry sweep

Version Date: 2026-06-11

ArchiveSourcesChat