Collaborative management of the Grand Ethiopian Renaissance Dam increases economic benefits and resilience

Freshwater and electricity are critical inputs to economic development and are tightly coupled across sectors. The Nile River, especially the Blue Nile originating in Ethiopia, has highly variable flows that underpin Egypt’s water, energy, food, and economic systems. The imminent completion of the Grand Ethiopian Renaissance Dam (GERD) presents opportunities (e.g., large hydropower capacity) and risks (e.g., altered downstream flows during filling/operation). Despite years of negotiation, Ethiopia, Sudan, and Egypt have not reached agreement on GERD’s filling and long-term operation. Prior analyses typically simplified linkages between hydrology and economies. This study asks whether a coordinated, information-sharing operating strategy for GERD can improve basin-wide water use and macroeconomic outcomes relative to a proposed operating approach resembling the 2019–2020 Washington draft proposal. The paper’s purposes are to present a dynamic, coevolutionary hydro–economic modeling framework coupling an Eastern Nile river system model with a computable general equilibrium (CGE) model of Egypt, and to use it to evaluate hydrological and economy-wide effects of coordinated GERD operation versus the Washington draft proposal. The importance lies in capturing feedbacks between hydrologic variability, infrastructure operation, and economy-wide performance for transboundary water management.

The paper situates its contribution within literature linking water and energy to economic growth, documenting bidirectional relationships between energy consumption and development and the interdependence of water–energy–food systems. In the Nile Basin context, earlier technical studies identified significant hydropower potential on the Blue Nile and highlighted management complexity during multiyear droughts. Negotiation milestones include the Nile Basin Initiative, International Panel of Experts recommendations, the 2015 Declaration of Principles, and subsequent technical processes (e.g., NISRG), none culminating in agreement. Previous GERD studies assessed filling and operations impacts on riparians but generally used simplified couplings between river systems and regional economies. Hydro-economic modeling reviews note approaches treating water either as explicit/implicit production factors or commodities in CGE models, and energy commonly combined with capital. This study advances the literature by explicitly coupling a detailed river operations model (Pywr) with an economy-wide CGE model for Egypt to capture two-way feedbacks and hydrologic stochasticity across multiple flow traces.

The study develops a coevolutionary hydro–economic framework linking: (1) an economy-wide dynamic-recursive CGE model of Egypt based on IFPRI’s standard model, extended to include water, energy, and land with nested production and consumption structures; and (2) a monthly-resolution Eastern Nile river system model implemented in Pywr that represents major dams (including GERD and High Aswan Dam, HAD), withdrawals, hydropower, non-hydro generation (aggregated), and network constraints. Integration and iteration are managed via the Python Pynsim framework, with the CGE (annual time step) and Pywr (monthly) models exchanging data each year until convergence (GDP threshold US$5 million; max 50 iterations). The CGE model: - Production uses a three-level nesting with Leontief at the top (composite intermediates and value-added–energy), CES between energy and value-added, and CES among energy commodities (including electricity). - Value-added combines labor, capital, and land via CES. - Four capital types are distinguished: hydro, non-hydro, municipal water (water capital), and general capital. Land and water capital utilization responds to rents; general and non-hydro capital accumulate via investment allocation by returns. Hydropower capacity is fixed over the 30-year horizon. - Households (10 groups by urban/rural and income quintile) use a nested LES–CES system for consumption. - Small open-economy assumption with fixed world prices; flexible exchange rate; fixed saving propensities; government spending as fixed share of total absorption. - Exogenous dynamics follow SSP “middle of the road” for labor and TFP. The river system model: - Uses naturalized inflows (1901–2002) from Eastern Nile Technical Regional Office; calibrated/validated at 8 locations (1970–2002) for flows and reservoir levels. - Represents Sudanese reservoirs (Roseires, Sennar, Merowe), GERD, HAD, irrigation and municipal demands, evaporation/spills (including Toshka), channel seepage. - Pywr allocates flows via linear programming with priorities and operating rules; recorders aggregate annual irrigation/municipal supply fractions and hydro/non-hydro generation for CGE shocks. Hydro–economic iteration: - CGE first estimates changes in irrigation/municipal water demand, electricity demand, and non-hydro capacity. - Pywr scales demands/capacity, simulates 12 months, returns annual supply fractions and generation. - CGE applies supply fractions as shocks to land and water capital; ratios of current to base-year hydro/non-hydro generation shock hydro/non-hydro capitals; re-solves. - Iterate until convergence or max iterations, then advance to next year. Scenarios and hydrology: - 102 synthetic 30-year flow traces (2020–2049) generated via Index Sequential Method over 1901–2002 flows. Operating strategies compared: (A) Washington draft proposal (assumptions reflecting 2019–2020 talks) versus (B) Coordinated operation. Washington draft proposal assumptions: - Initial filling (5-year plan per Supplementary Table 3): retain inflows in July–August to targets (year 1: 4.9 bcm, enabling 2×375 MW turbines; year 2: 18.4 bcm enabling remaining turbines); maintain minimum environmental release 43 Mm³/day; Sept–June releases equal inflows; allow filling delays during drought. - Long-term operation: begins at storage ≥49.3 bcm; target monthly energy 1170 GWh (90% reliability); reduced to 585 GWh if storage <49.3 bcm; maintain 43 Mm³/day environmental release; drought mitigation mechanisms: (i) annual inflow ≤37 bcm triggers minimum annual release (dependent on storage/inflow), (ii) 4-year average release ≥39 bcm (≥37 bcm during initial filling), (iii) 5-year average release ≥40 bcm; implemented by checking in March each hydrologic year and adjusting March–May releases to meet thresholds. Coordinated operation assumptions: - Initial filling similar targets but not restricted to July–August; maintain 43 Mm³/day environmental release. - Core rule: when physically possible, GERD monthly releases ≥ (Sudan’s Blue/Main Nile withdrawal targets) + (Egypt’s HAD target release when HADR <50 bcm, equivalent to 156 m asl). When HADR ≥50 bcm, Ethiopia has more flexibility for storage/hydropower. - Sudanese dams’ operations adapted to pass GERD releases intended for Egypt. - Long-term operation also targets 1170 GWh monthly at storage ≥49.3 bcm (585 GWh below), with the added coordinated releases rule above; no separate fixed minimum annual release drought mechanisms—dynamic deficit-reduction triggers instead. Economic valuation: - Egypt’s macroeconomic impacts computed endogenously by CGE. - GERD electricity value for Ethiopia assessed via present value using export price US$0.05/kWh at 3% discount rate (without an Ethiopian CGE due to data/market allocation uncertainties). Simulation horizon is 2020–2049 for both components.

- Water supply and losses: • In 77% of traces, coordinated operation reduces Egypt’s total irrigation water deficits versus the Washington draft. Significant reductions occur mainly during multiyear scarcity periods after 2025 when HADR <50 bcm. • Total water losses (evaporation from reservoirs, Toshka spills, channel seepage) over 30 years change by −18.1 to +1.1 bcm (median −5.1 bcm) under coordinated operation. • 30-year accumulated change in Egypt’s irrigation withdrawals: max +27.3 bcm; 90th percentile +16.2; median +1.4; 10th percentile 0.0; min −1.7 bcm. - Hydropower: • GERD cumulative generation increases in 71% of traces; 2020–2049 change ranges from −1,900 to +17,900 GWh (median +1,600 GWh). Annual GERD generation changes range from −35% to +84% (median 0%, increases in 40% of years, decreases in 10%). • Egypt’s hydropower generation often declines early (2020–2030) due to faster GERD filling and HADR drawdown; over 30 years, accumulated change in Egypt hydropower: max +2.7; 90th +1.2; median −0.1; 10th −1.0; min −2.1 (thousand GWh). • System-wide hydropower generally increases in later years due to higher GERD storage and opportunistic coordinated management. - Macroeconomic outcomes (Egypt, present values, 3% discount): • GDP: change ranges −US$0.7 to +US$4.1 billion; median +US$0.24 billion; increases in ~76% of traces. • Investment, exports, imports, and government savings all tend to increase with median PV changes of roughly US$80M, US$70M, US$70M, and US$20M, respectively. Improvements concentrate in 5–15 drought years when HADR <50 bcm. - Ethiopia’s GERD electricity value (at US$0.05/kWh, 3% discount): • Present value change ranges −US$0.05 to +US$0.60 billion; median +US$0.06 billion; increases in 71% of traces. - Mechanistic insights: • Coordinated operation accelerates GERD filling when HADR is sufficiently full, shifts some storage upstream (lower HADR levels, higher GERD levels in some traces), reduces downstream water deficits during scarcity via dynamic releases, and reduces basin losses. • Sudan’s modeled benefits (irrigation reliability, hydropower, flood control) are essentially similar under both strategies given daily coordination and data sharing, so detailed Sudan outcomes are not separately reported.

The findings address the central question of whether coordination improves basin-wide outcomes: coordinated operation reduces Egypt’s irrigation deficits in most hydrologic sequences, lowers basin water losses, and increases GERD and system-wide hydropower across a majority of traces, thereby enhancing economic resilience in Egypt and providing additional electricity value to Ethiopia. Importantly, benefits accrue during multiyear droughts when cooperation is most needed, highlighting the value of dynamic, information-sharing rules that link GERD releases to downstream conditions (e.g., HADR <50 bcm) rather than fixed annual minimums. By explicitly modeling hydrologic stochasticity and two-way feedbacks between water–energy supply and the macroeconomy, the analysis shows that infrastructure operating strategies can have economy-wide repercussions beyond direct sectoral metrics. The coordinated strategy offers a technically flexible path that can build trust and align incentives, potentially facilitating broader regional integration in water–energy–food systems. The results underscore the importance of agreements that are legally binding yet adaptable to future development and allocation arrangements.

The paper contributes a coevolutionary hydro–economic modeling framework that couples a detailed river system simulator with a dynamic CGE model to capture feedbacks between hydrology, infrastructure operations, and the macroeconomy. Applying this framework to GERD shows that a coordinated, information-sharing operating strategy can: (a) reduce Egypt’s irrigation water deficits in most scenarios; (b) decrease basin water losses; (c) raise GERD and system-wide hydropower in most traces; and (d) improve Egypt’s macroeconomic indicators and Ethiopia’s electricity value relative to a Washington draft-like operation. Policy-wise, adopting a technically flexible, legally amendable coordinated strategy could foster trust and pave the way for deeper regional integration. Future research should extend the framework to include: (i) explicit economy-wide modeling for Ethiopia and Sudan; (ii) environmental and social impacts not captured by GDP; and (iii) prospective irrigation expansions and other infrastructural developments across the basin to evaluate long-term allocation and resilience trade-offs.

- The economy-wide analysis focuses on Egypt; Ethiopia’s macroeconomic impacts from GERD hydropower are not modeled endogenously due to uncertainties in domestic consumption, export volumes, and prices (GERD electricity value is proxied using a fixed export price of US$0.05/kWh). - Potential future irrigation expansion in Ethiopia, Sudan, and South Sudan (estimated at ~2.4 million ha) is not modeled; such expansion would likely reduce flows to Egypt. - Environmental and social impacts beyond macroeconomic indicators (e.g., biodiversity, local livelihoods such as recession agriculture) are not captured in GDP-based results and are assumed similar across the two operating strategies for Sudan. - Results depend on assumptions about operating rules (e.g., HADR 50 bcm threshold), drought mitigation implementations, and model calibrations; uncertainties in hydrologic forecasts are represented via historical flow traces but not climate change projections. - Data availability constraints limited public access to some river system inputs; calibration/validation rely on 1970–2002 observations.