Côte d’Ivoire is located in West Africa, along the Gulf of Guinea, between 2˚30' and 8˚30'W and 4˚30' and 10˚30'N (Figure 1). The country covers an approximately square territory with a 515 km Atlantic coastline and land borders with Liberia, Guinea, Mali, Burkina Faso, and Ghana.

According to the 2021 Population and Housing Census (RGPH 2021), Côte d’Ivoire had a resident population of 29,389,150 inhabitants [3]. The country is characterized by substantial demographic growth and a sizeable non-Ivorian population, which accounted for 24.2% of the total population in the 2014 census [2].

Figure 1. Geographical location of Côte d’Ivoire, West Africa, using QGIS. (a) Côte d’Ivoire, with Abidjan and Yamoussoukro; (b) location of Côte d’Ivoire in Africa. Boundaries: Natural Earth 1:10 m Admin 0 (public domain).

Côte d’Ivoire’s economy has expanded steadily since 2012. Real GDP growth reached 7.62% in 2012, 10.76% in 2013, 9.37% in 2014, 7.19% in 2015, 7.17% in 2016, and 7.41% in 2017 [5] (Figure 2). Growth subsequently moderated to 4.84% in 2018 and 6.72% in 2019, before slowing sharply to 0.70% in 2020 during the COVID-19 pandemic. The economy rebounded in 2021, with real GDP growth reaching 7.06% [5]. Overall, real GDP growth averaged approximately 6.9% per year between 2012 and 2021. Côte d’Ivoire remains the largest economy within the West African Economic and Monetary Union (WAEMU) according to the African Development Bank’s 2026 country assessment [13].

Figure 2. Evolution of real GDP growth in Côte d’Ivoire, 2000-2021. Source: World Bank, World Development Indicators (WDI), GDP growth (annual %) [5].

2.3.1. Energy Demand Model

Future electricity demand was projected using the Model for Analysis of Energy Demand (MAED-2), developed by the International Atomic Energy Agency (IAEA) and based on the end-use methodology originally formulated at the Institut Économique et Juridique de l’Énergie (IEJE), University of Grenoble, under the name MEDEE-2. MAED-2 is a scenario-based planning tool in which each scenario represents a coherent long-term development pathway linking energy demand to demographic, socio-economic, and technological driving factors [14], [15]. Its end-use framework enables the representation of structural changes in energy-consuming activities, technological development, and substitution among energy forms over the projection horizon.

Within MAED-2, energy demand is disaggregated by sector and end-use activity and linked to relevant demographic, socio-economic, and technological drivers. In the present application, final demand was structured across four main consumer sectors: industry (including agriculture, mining, manufacturing, and construction), household, service, and transport (Figure 3).

2.3.2. Observational Data

Historical inputs comprised demographic indicators (urban and rural population and population growth), labour-force statistics, macroeconomic indicators (GDP

Figure 3. MAED-2 modelling procedure, from base-year data and development driving forces to scenario-specific projected electricity demand, disaggregated by end-use sector.

and sectoral composition), sector-specific energy intensities, household end-use parameters, and base-year electricity data. Data were compiled from the Institut National de la Statistique, World Bank indicators, and national electricity-sector statistics. The primary source for the 2013 electricity base year was the ANARE-CI 2013 energy balance. The MAED-2 base value of 6267.5 GWh is the electricity supplied to the domestic distribution network (7518.5 GWh entering the transmission network, minus 457.6 GWh of transmission losses, 789.0 GWh of net exports, and 4.4 GWh supplied to direct customers); exports are therefore excluded from the base-year quantity. Gross electricity production, exports, electricity supplied to the distribution network, distribution losses, and final billed consumption were treated as distinct electricity-accounting quantities rather than interchangeable measures [16]. Because this boundary lies upstream of distribution losses (1233.9 GWh in 2013, i.e. 19.7% of the base value [16]), the MAED-2 base implicitly includes the 2013 distribution losses; this rate subsequently fell to 13.0% in 2020, 8.9% in 2023 and 10.0% in 2024 [17], so that part of the retrospective gap reflects reduced distribution losses rather than lower final demand (Section 4.4).

2.3.3. Modelling of Future Electricity Consumption

The modelling procedure began with construction of the 2013 base-year energy-demand structure and reconciliation of the demographic, socio-economic, technological, and electricity-sector inputs required by MAED-2. Four internally consistent scenario assumption sets were then developed, structured around two complementary components: 1) a socio-economic component describing alternative demographic and economic trajectories, and 2) a technological component representing changes in end-use efficiency and the penetration of alternative energy forms. MAED-2 was subsequently executed for the Low (LS), Reference (RS), High (HS), and Optimistic (OS) scenarios. Electricity-demand outputs were obtained directly from MAED-2 for the model years 2013, 2020, 2025, 2030, 2035, 2040, and 2045.

The modelling sequence (Figure 3) comprised five steps: 1) disaggregation of total final energy demand by consumption sector, fuel type, and energy form; 2) identification of the socio-economic and technological drivers associated with each end-use category; 3) reconstruction of base-year (2013) final energy consumption from available statistical data; 4) construction of development scenarios reflecting alternative demographic, macroeconomic, and technological trajectories; and 5) evaluation of projected energy demand for each scenario over the 2013-2045 horizon.

For the retrospective year-specific comparison, direct MAED-2 outputs were available for 2020 and 2025, whereas outputs for the intervening years 2021-2024 were not available. Scenario values for these four years were therefore obtained by local exponential interpolation between the corresponding direct 2020 and 2025 MAED-2 outputs:

with E D s , y is the interpolated electricity demand for scenario s in year y; E D s , 2020 and E D s , 2025 are the corresponding direct MAED-2 outputs.

This interpolation was used only to obtain year-specific values for the retrospective comparison and was not used to generate the MAED-2 model-year projections or the long-term scenario trajectories. The observational series used for the retrospective stress test is presented in Section 3.3 using a consistent distribution-network accounting boundary wherever explicitly available in ANARE-CI annual energy balances.

2.3.4. Retrospective Assessment and Statistical Metrics

The retrospective assessment compared the MAED-2 scenario trajectories with observed distribution-network supply over 2020-2024. The differences between the scenario values and observed distribution-network supply were quantified using the mean bias error (MBE) and the mean absolute percentage error (MAPE):

where P i represents the MAED-2 scenario value, O i the observed distribution-network supply, and n the number of years included in the retrospective comparison. MBE indicates the direction and average magnitude of the deviation, with positive values indicating that the scenario values exceed the observations and negative values indicating the opposite. MAPE expresses the average absolute deviation relative to the observed values as a percentage. Because the scenario values exceeded the observations in every year, the mean absolute error equals the MBE and is not reported separately.

The two metrics were calculated separately for LS, RS, HS, and OS over the 2020-2024 comparison period. They were used to quantify the deviation between each scenario trajectory and the observed distribution-network supply rather than as measures of formal model validation.

Four development scenarios were constructed for Côte d’Ivoire: The Low Scenario (LS), Reference Scenario (RS), High Scenario (HS), and Optimistic Scenario (OS), characterized by average GDP growth rates over the projection period of 7%, 9%, 10%, and 11%, respectively. These values represent average growth assumptions over the projection horizon rather than constant annual growth rates. The underlying model-year GDP growth assumptions vary over time in all four scenarios and are reported in the Supplementary Material. Beyond GDP growth, the four scenarios also differ in demographic evolution, sectoral development, technological change, energy-efficiency assumptions, and broader development conditions.

The numerical MAED-2 scenario inputs relevant to the electricity-demand analysis for the model years 2013, 2020, 2025, 2030, 2035, 2040, and 2045 are provided in the Supplementary Material. The supplementary tables report population and household parameters, urbanization, electrification-related assumptions, sectoral activity and economic structure, electricity-specific end-use intensities, electricity penetration/substitution assumptions, transport assumptions relevant to electricity demand, and electricity-related end-use efficiency parameters. Consistent with the scope of this study, parameters associated exclusively with non-electric energy carriers are not reported as independent electricity-demand assumptions. Formula-linked or derived MAED-2 quantities are distinguished from user-entered scenario assumptions where applicable. Table 1 summarises the main qualitative assumptions distinguishing the four development scenarios, with further details provided in Sections 2.4.1-2.4.4. The detailed numerical inputs, including the time-varying model-year GDP growth assumptions of the four scenarios, are reported in the Supplementary Material (Tables S1-S9).

2.4.1. Reference Scenario

The Reference Scenario (RS) is characterized by an average GDP growth rate of 9% over the projection period. The underlying model-year GDP growth assumptions vary over time. Demographically, the RS incorporates time-varying population-growth and household-size assumptions. Urban household size decreases from 5.00 persons per household in 2013 to 4.50 in 2045, while rural household size decreases from 5.70 to 5.50; the urban population share increases from 49.5% to 55.0%. Sectorally, the RS assumes relatively balanced development across economic sectors, without a single dominant driving sector. Technologically, energy efficiency improves in line with historical trends, without accelerated adoption of energy-efficient technologies.

Table 1. Summary of demographic, sectoral, and technological assumptions underlying the four development scenarios.

2.4.2. Low-Growth Scenario (LS)

The Low Scenario (LS), labelled the 7% scenario, uses time-varying GDP growth assumptions of 8.0%, 7.3%, 7.1%, 6.5%, 6.9%, and 7.0% for 2020, 2025, 2030, 2035, 2040, and 2045, respectively. The scenario reflects reduced domestic and foreign investment under conditions of regional political instability and potential climatic or economic shocks. Demographically, the LS does not assume a constant household size: urban household size decreases from 5.00 persons per household in 2013 to 4.50 in 2045, while rural household size varies from 5.70 to 5.74 over the same endpoints. The urban population share increases from 49.5% to 55.0%. Sectorally, industrial development remains limited, with the mining and manufacturing sectors expanding only marginally over the study period.

2.4.3. High-Growth Scenario (HS)

The High Scenario (HS) is characterized by an average GDP growth rate of 10% over the projection period. The underlying model year GDP growth assumptions vary over time. The scenario reflects increased private and public investment and accelerated industrialization. Demographically, urban household size decreases from 5.00 persons per household in 2013 to 4.50 in 2045, while rural household size changes from 5.70 to 5.74. The urban population share increases from 49.5% to 55.0%. Sectorally, growth is led by industry, particularly through the expansion of manufacturing and mining activities, alongside continued development in other economic sectors.

2.4.4. Optimistic Scenario (OS)

The Optimistic Scenario (OS) is characterized by an average GDP growth rate of 11% over the projection period. The underlying annual/model-year GDP growth assumptions vary over time. The scenario reflects strong simultaneous development across economic sectors. Demographically, the OS uses time-varying household-size assumptions: urban household size changes from 5.00 persons per household in 2013 to 4.50 in 2045 and rural household size from 5.70 to 5.10. The urban population share increases from 49.5% to 55.0%. Sectorally, growth is led by industry and manufacturing, alongside strong development in other economic sectors. Technologically, the OS assumes rapid adoption of energy efficient technologies and improvements in industrial processes, supported by sustained investment in modernisation.

The year 2013 was selected as the MAED-2 base year to establish the demographic, economic, and electricity-demand conditions used for model initialization. Subsequent model years were defined as 2020, 2025, 2030, 2035, 2040, and 2045.

Base-year economic activity was characterized using sectoral value-added indicators from the World Bank’s World Development Indicators (WDI). In 2013, agriculture, forestry and fishing accounted for 15.34% of GDP, manufacturing for 11.85%, and services for 38.81%, while industry including construction represented 18.98% of GDP. Because manufacturing is included within the WDI industry aggregate, these indicators were not treated as additive components. The industry aggregate was used to characterize the overall industrial structure, while the detailed sectoral assumptions were specified separately for the subsequent scenario construction [18].

Demographic, labour-market, and household conditions for the base year were compiled from official national statistics. According to the 2013 employment survey conducted by INS Côte d’Ivoire, the projected national population was 23.203 million, of which 14.501 million were of working age, corresponding to 62.5% of the total population. Household characteristics were estimated at 5.756 million households nationwide, with an average household size of 4.0 persons per household [19]. The demographic and household parameters used for the base-year reconstruction are summarized in Table 2. These INS values are statistical references; in the MAED-2 workbooks, the 2013 population is initialised at 22.106 million and household size is specified separately for urban (5.00) and rural (5.70) households, with an urban population share of 49.5% (Supplementary Material, Table S4).

Table 2. Demographic and household parameters for the 2013 base-year reconstruction [19].

Table 2. Demographic and lifestyle parameters for the 2013 base year.

MAED-2 projects a sustained increase in electricity demand across all four scenarios over the 2013-2045 period (Figure 4; Table 3). The trajectories are based on direct MAED-2 outputs for the model years 2013, 2020, 2025, 2030, 2035, 2040, and 2045. Starting from the common 2013 base-year value of 6267.5 GWh (exact model value: 6267.459 GWh), electricity demand increases progressively over the projection horizon under each scenario.

In 2020, projected demand reaches 11589.4 GWh under LS, 11462.8 GWh under RS, 12202.7 GWh under HS, and 13633.2 GWh under OS. The LS value for 2020 exceeds the RS value despite lower GDP growth, because its services-sector electricity demand (5158 vs 4450 GWh) more than offsets its lower industrial demand (4547 vs 5138 GWh). By 2030, the corresponding values increase to 24466.9, 27010.5, 35207.4, and 41355.2 GWh, respectively. The upward trajectories continue through 2040, when projected demand reaches 51346.3 GWh under LS, 68715.2 GWh under RS, 110121.3 GWh under HS, and 148439.9 GWh under OS. By 2045, demand reaches 77755.1 GWh under LS, 112978.9 GWh under RS, 176554.3 GWh under HS, and 295542.7 GWh under OS.

Over the full 2013-2045 projection period, the corresponding endpoint-implied compound annual growth rates are 8.19%, 9.46%, 11.00%, and 12.80% for LS, RS, HS, and OS, respectively.

The magnitude of the projected increase differs substantially across the four scenarios. Relative to the common 2013 base value, electricity demand in 2045 is approximately 12.4 times the base-year level under LS, 18.0 times under RS, 28.2 times under HS, and 47.2 times under OS. The resulting spread is particularly pronounced between the two outer scenarios, with the OS projection approximately 3.80 times the LS projection in 2045.

Table 3. Summary of projected electricity demand by scenario (2013 and 2045) and corresponding compound annual growth rates (CAGR).

Substantial differences are also observed between the intermediate scenarios. In 2045, projected electricity demand under HS is approximately 56.3% higher than under RS. Thus, considerable differences in long-term electricity-demand outcomes emerge not only between the lowest and highest scenarios but also between the intermediate development pathways.

These differences reflect the combined influence of the demographic, economic, sectoral, and technological assumptions embedded in each MAED-2 scenario. Although LS, RS, HS, and OS are characterized by average GDP growth rates over the projection period of 7%, 9%, 10%, and 11%, respectively, the differences in projected electricity demand cannot be attributed to GDP growth alone because several scenario assumptions vary simultaneously.

A retrospective comparison was conducted to assess how the MAED-2 electricity-demand pathways relate to the observed evolution of electricity supplied to the domestic distribution network after the 2013 base year. Figure 4 compares the four scenario trajectories with available distribution-network supply observations through 2024, while Table 4 reports the values used for the retrospective assessment. The observed series increased from the common base value of 6267.5 GWh in 2013 to 9245.2 GWh in 2020 and then rose every year, to 9918.5 GWh in 2021, 10541.3 GWh in 2022, 11580.6 GWh in 2023, and 12633.4 GWh in 2024 [17].

In 2020, the direct MAED-2 outputs were 11589.4 GWh for LS, 11462.8 GWh for RS, 12202.7 GWh for HS, and 13633.2 GWh for OS, compared with an observed distribution-network supply of 9245.2 GWh. The observed value was therefore below all four scenario outputs. This separation persisted through 2024, with the observed series remaining below the corresponding scenario trajectories (Figure 4).

Table 4. MAED-2 scenario trajectories and observed distribution-network electricity supply in Côte d’Ivoire, 2020-2024.

Note: Values marked with an asterisk for 2021-2024 were obtained by local exponential interpolation between the direct 2020 and 2025 MAED-2 outputs and are used solely for the retrospective comparison. Observed values: ANARE-CI annual reports [17] (energy balances; 2020: pp. 34-35; 2021: pp. 34-35; 2022: pp. 38-39; 2023: pp. 42-43; 2024: p. 51).

Figure 4. Comparison of MAED-2 electricity-demand trajectories with observed distribution-network supply in Côte d’Ivoire over 2013-2045. Observed distribution-network supply is shown for available years through 2024.

The differences between the MAED-2 scenario trajectories and observed distribution-network supply over 2020-2024 were quantified using the MBE and MAPE metrics defined in Section 2.3.4 (Table 5).

Table 5. Statistical metrics for the retrospective comparison between MAED-2 scenario trajectories and observed distribution-network supply over 2020-2024.

All four scenarios overestimated observed supply in every year (positive MBE), and the deviation increased from LS to OS. The supply observed in 2024 (12633.4 GWh) lies between the LS values for 2021 and 2022 (Table 4), i.e. roughly three years behind the lowest scenario pathway.

In sum, the retrospective comparison indicates that MAED-2 represents the long-term direction of increasing electricity demand, while the magnitude of projected demand varies substantially with the scenario assumptions.

The compound annual growth rates projected for Côte d’Ivoire (8.19% - 12.80%) are broadly comparable to those obtained in comparable end-use modelling studies elsewhere in West Africa. Using the same MAED framework, Kanté et al. [20] projected electricity demand growth rates of 8.13% (low scenario), 10.31% (reference scenario), and 12.56% (high scenario) for the Taoussa region of northern Mali over 2020-2035 a range similar to that obtained in the present study, despite differences in country, study period, and spatial scale (national vs. sub-national). This similarity across two independent West African applications of the MAED methodology provides a useful regional plausibility context for the magnitude of the projected growth rates.

A more recent MAED-D application to Sierra Leone [21], one of the least electrified countries in sub-Saharan Africa, similarly projected substantial relative demand growth across baseline, high-, and low-demand scenarios to 2050, from an extremely low per-capita consumption base. Although expressed in different units (petajoules rather than gigawatt-hours) and therefore not directly comparable in absolute terms, this study reinforces the broader regional pattern of rapid projected electrification-driven demand growth across data-constrained sub-Saharan African economies.

MAED-2 has also been applied to long-term energy- and electricity-demand analysis in Asian developing economies, including Bangladesh [6] and Syria [7], which supports its methodological relevance beyond the West African context.

The magnitude of electricity demand growth projected for Côte d’Ivoire contrasts sharply with long-term projections for mature, highly electrified economies. In the European Union, electricity demand is projected to grow at an average annual rate of approximately 1.1% between 2020 and 2050 [22]. More recent short-term projections indicate that electricity demand increased by 0.9% in 2025 and is expected to grow by approximately 2.3% per year over 2026-2030, driven by increasing electricity use in buildings, transport, industry, and data centres [23]. In the United States, electricity demand increased by 2.1% in 2025 and is projected to grow by close to 2% per year on average through 2030, with data-centre expansion expected to account for approximately half of the increase [23]. The long-term EIA reference case projects growth of approximately 0.9% - 1.6% per year through 2050 [24].

These growth rates remain substantially below the 8.19% - 12.80% endpoint-implied CAGR obtained for Côte d’Ivoire, reflecting different stages of economic and energy-system development. In mature economies, recent electricity-demand growth follows a prolonged period of relatively stagnant demand associated partly with efficiency improvements and industrial restructuring, whereas Côte d’Ivoire is characterized by a lower electricity-consumption base and continuing demographic and economic development. This contrast is consistent with the broader global pattern reported by the International Energy Agency: emerging markets and developing economies accounted for approximately 80% of global electricity-demand growth in 2025 and are expected to maintain a similar share through 2030, compared with an average of approximately 95% over the previous decade. The declining share reflects the renewed contribution of advanced economies, where electricity demand is increasing again after approximately 15 years of stagnation, supported by data centres, electrification, advanced manufacturing, and other emerging loads [23].

This comparative analysis situates the present projections for Côte d’Ivoire within two complementary bodies of evidence. First, the similarity with the Malian case study [20], using the same modelling framework in a neighbouring West African country, provides a useful regional benchmark for the magnitude of projected demand growth. Second, the marked contrast with mature-economy trajectories in Europe and the United States [22]-[24] places the Côte d’Ivoire projections within a different demographic, economic, and electricity-sector development context, in which continued growth in electricity use has important implications for future generation capacity and grid infrastructure. This comparative perspective reinforces the practical relevance of the scenario-based approach adopted in this study for examining alternative long-term electricity-demand pathways and informing electricity-system planning in Côte d’Ivoire.

A useful test of scenario relevance is whether subsequently observed outcomes remain within, or systematically depart from, the scenario envelope. The retrospective comparison presented in Section 3.3 shows that observed distribution-network supply remained below all four MAED-2 scenario trajectories over 2020-2024. Among the four scenarios, the Low Scenario (LS) showed the smallest deviation from the observed series, although observed distribution-network supply remained below this trajectory throughout the comparison period.

The quantitative assessment further illustrates this divergence. Over 2020-2024, LS produced the lowest MBE and MAPE among the four scenarios, followed by RS, HS, and OS (Table 5). The increasing magnitude of the deviations toward the higher-demand scenarios indicates that the observed evolution during the early part of the projection horizon has been closer to the lower end of the scenario range than to the higher-demand pathways. This finding does not invalidate the longer-term scenarios, which represent alternative development pathways through 2045 rather than short-term forecasts.

Two factors explain much of this divergence. First, distribution losses fell sharply after 2013: at the 2013 loss rate (19.7%), the electricity delivered to consumers in 2024 would have required about 14,160 GWh of distribution-network supply instead of 12633.4 GWh, which corresponds to roughly two-thirds of the 2024 deviation from LS. Second, economic growth during the retrospective period did not consistently follow the scenario assumptions; because the scenarios combine several demographic, economic, sectoral, and technological assumptions, the remaining gap cannot be attributed to GDP growth alone.

Under the 2014 Electricity Code [11], ANARE-CI proposes electricity tariffs to the State, a mechanism applied to the 10% tariff adjustment effective from 1 January 2024 [25]. The present analysis does not model sector revenues, costs, or tariff formation; its implication is that demand assumptions constitute an important input to electricity-sector planning and that comparing them with subsequently observed data strengthens the evidence available to planners and regulators.

The results are also relevant to investment planning. Côte d’Ivoire’s Pacte National de l’Énergie, published in January 2025 in connection with the World Bank Group and African Development Bank Mission 300 initiative [12], estimates total financing needs of USD 15.02 billion, of which USD 6.56 billion remained to be financed, and aims to mobilise about USD 2 billion of private investment in electricity generation, transmission, and distribution. The national electricity-sector investment plan to 2040 estimates average annual growth in domestic electricity consumption at 6.5% over 2025-2030 [26]. Over the same 2025-2030 period, the direct MAED-2 outputs imply average growth of 8.9% (LS), 9.7% (RS), 11.7% (HS), and 10.9% (OS) per year, all above this benchmark; the lower OS than HS rate over this sub-period mainly reflects slower growth of industrial (12.1% vs 13.6% per year) and household (1.1% vs 2.9% per year) electricity demand in OS. This comparison indicates that the MAED-2 scenarios embed stronger demand-growth assumptions than the recent national planning reference and provides a contemporary benchmark for interpreting the magnitude of the scenario trajectories.

More broadly, the substantial differences among the LS, RS, HS, and OS trajectories indicate that future generation and network requirements may vary considerably depending on the evolution of electricity demand. Transparent scenario envelopes combined with retrospective comparison can therefore help planners and investors assess the implications of alternative demand pathways for generation-capacity expansion and transmission and distribution infrastructure. The results support the use of alternative demand scenarios rather than reliance on a single long-term demand trajectory when examining future electricity-system requirements.

This study has several methodological strengths. First, the four electricity-demand scenarios were constructed and executed using the MAED-2 end-use framework, allowing long-term demand to be represented under alternative demographic, economic, sectoral, and technological development pathways. Second, the analysis uses direct MAED-2 outputs for the model years 2013, 2020, 2025, 2030, 2035, 2040, and 2045 rather than reconstructing the long-term scenario trajectories from the 2013 and 2045 endpoints. Third, the retrospective comparison with observed distribution-network supply through 2024 provides an additional assessment of how the early evolution of the electricity system relates to the original scenario pathways. The use of MBE and MAPE further provides a quantitative characterization of these deviations.

Several considerations are important when interpreting these results. The four scenarios represent alternative combinations of demographic, economic, sectoral, and technological assumptions rather than probabilistic forecasts. Because these drivers vary jointly across scenarios, the relationship between GDP growth and electricity-demand growth should not be interpreted as an isolated GDP effect. In addition, observed distribution-network supply does not necessarily represent unconstrained electricity demand. The 2020-2024 retrospective comparison is therefore used to assess the early alignment of observed electricity-system evolution with the original scenario trajectories.

The combination of end-use scenario modelling, transparent scenario assumptions, direct model outputs, and retrospective comparison provides a coherent framework for examining alternative long-term electricity-demand pathways in Côte d’Ivoire. These demand pathways can also provide inputs for subsequent assessments of generation, network, and flexibility requirements.