Progress of the Malabo Declaration as a Regional Agenda Towards Addressing Hunger in Africa

Abstract

The Malabo Declaration commits African Union member states to eliminating hunger by 2025. Progress toward this target has been uneven and poorly understood. While some countries have recorded gains in non-hunger thematic areas such as finance, trade, resilience to climate variability, and governance and accountability mechanisms, the extent to which these improvements contribute to hunger reduction remains unclear. This study investigates whether performance in non-hunger areas, as measured through the Comprehensive Africa Agriculture Development Programme Biennial Review C-scores, is statistically associated with outcomes under Commitment 3, which focuses on hunger reduction. We used random effects panel regression model covering 55 African countries from 2017 to 2023, the analysis identifies five significant predictors: agricultural GDP and poverty reduction (PC 4.1), foreign private investment (PC 2.3), multi stakeholder coordination (PC 1.2), inclusive public–private partnerships (PC 4.2), and trade policies (PC 5.2). Investment in resilience (PC 6.2) and capacity for planning and monitoring (PC 7.1) showed marginal associations. Our findings suggest that institutional presence alone does not drive hunger outcomes. We reflect that what matters is the structure, inclusiveness, and functionality of these mechanisms, including whether investments reach food-insecure populations, coordination platforms influence decisions, and policies adapt to local conditions. This study concludes that some high-performing categories fail to deliver tangible hunger reduction benefits when implementation is fragmented or disconnected from context. These findings challenge how progress is currently measured and interpreted at the regional level. Finally, we reiterate that as the region prepares for the post-2025 agenda, future strategies must directly link agricultural transformation to hunger reduction through targeted interventions and accountable institutions.

1. Introduction

Food security remains one of Africa’s critical development priorities amid rapid population growth [1,2]. Despite significant policy shifts toward agriculture, especially following the limited success of the Green Revolution, many reforms across the region have failed to achieve the sustained agricultural growth needed to reduce poverty and hunger [3]. In recent times, economic instability, climate change, and conflict are increasingly destabilising food systems [4,5]. These interconnected crises create feedback loops in which economic shocks disrupt food supply chains, climate variability intensifies agricultural vulnerabilities, and conflict leads to displacement and resource scarcity [4,5,6,7].

As of 2023, 298.4 million people in Africa were undernourished, accounting for approximately 40% of the global total, which ranged between 713 and 757 million people [8]. Africa also had the highest prevalence of undernourishment at 20.4%, compared to 8.1% in Asia and 6.2% in Latin America and the Caribbean. Although Asia had more undernourished people in absolute terms (384.5 million) due to its larger population, hunger was more widespread in Africa when measured as a share of its population [8]. A study by Staatz et al. [9] noted that agricultural output alone cannot guarantee hunger reduction where inequality, poor access, and exclusion persist. Recent analyses by Ulimwengu et al. [10] call for a shift from production-focused strategies to food systems approaches that embed resilience, equity, and accountability. These perspectives mirror the layered structure of the Malabo framework, which spans upstream drivers like governance and finance, midstream actions such as investment and productivity, and downstream outcomes including nutrition and poverty [11]. However, the actual alignment between these layers remains underexplored.

Overview of the Regional Agenda Towards Addressing Hunger in Africa

Across Africa, the commitment to eliminate hunger by 2025 is not on track, with significant disparities in progress across regions and countries. The African Union adopted the Malabo Declaration in 2014 in response to these challenges (see Figure 1). Building on the 2003 Comprehensive African Agricultural Development Programme (CAADP), the Declaration outlined seven commitments intended to transform African agriculture and achieve food security. Commitment 3 focuses specifically on ending hunger by 2025 through doubling productivity, reducing post-harvest losses, and improving nutrition [12]. The Declaration outlines ambitious targets for agricultural transformation and hunger reduction, yet most regions and countries are falling short of these goals [13,14,15,16]. The six performance categories (PC) used to evaluate Commitment 3, including access to inputs and technologies (PC 3.1), agricultural productivity (PC 3.2), post-harvest loss reduction (PC 3.3), social protection (PC 3.4), food and nutrition security (PC 3.5), and sanitary and phytosanitary measures (PC 3.6), have recorded only limited gains [16,17].

The available data suggest that progress toward these targets remains limited. With the deadline approaching, it appears unlikely that the region will meet its hunger eradication goals. Policy evaluations attribute this gap to weak implementation capacity, poor alignment between national and regional strategies, and inadequate cross-sector coordination [10,17]. Others highlight limitations in the monitoring framework, such as data inconsistency, indicator sensitivity, and scoring systems that may obscure real progress [11]. Beyond the technical and institutional issues, the underlying logic of the Malabo framework requires critical reassessment. Although Commitment 3 focuses specifically on hunger reduction, it is embedded within a broader transformation agenda that spans governance, finance, trade, poverty reduction, resilience, and accountability mechanisms. These are presumed to act as enablers of food security and improved livelihoods [12,18]. However, whether progress in these thematic areas has contributed to hunger reduction has not been empirically validated. The framework assumes synergy, but it remains unclear whether it functions as an integrated system or a collection of siloed targets.

In addition to capacity constraints, investment priorities are interpreted, such as whether they strengthen local food systems or prioritise export-driven value chains, which also affects how upstream and midstream efforts influence hunger outcomes [19,20]. There is limited empirical evidence on how these components interact to support Commitment 3. The relationships across performance categories and the extent to which gains in non-hunger areas support hunger reduction are yet to be clearly established.

This study addresses that gap by first assessing regional disparities in hunger reduction using T-scores and C-scores from the 2017 to 2023 CAADP Biennial Reviews. It then applies a country-level panel regression model to test whether performance in non-hunger thematic areas is statistically associated with outcomes under Commitment 3, and examines how these relationships vary across countries and sub-regions in Africa.

2. Methodology

This study employs a panel regression framework to assess the relationship between non-hunger thematic performance and hunger reduction outcomes across 55 African countries, drawing on data from four CAADP Biennial Review cycles (2017, 2019, 2021, 2023). The dependent variable, Commitment 3 (hunger reduction) T-scores, measures national progress toward eliminating hunger on a standardised 0–10 scale, benchmarked against annual milestones derived from linear progress expectations toward 2025 targets. Independent variables include 18 Performance Category (C) scores from non-hunger thematic areas (Commitments 1, 2, 4–7), such as agricultural investment, trade, and institutional capacity, each aggregated from indicator-level scores and scaled 0–10.

The analysis utilises a linear panel regression model to account for cross-country heterogeneity. Both Fixed Effects (FE) and Random Effects (RE) specifications were considered, with model selection guided by the Hausman test. The test’s p-value of 1.000 indicated no significant difference between FE and RE coefficients, supporting the use of the RE model for its efficiency in estimating time-invariant factors. An OLS-based RE estimator with unadjusted standard errors was applied, excluding time fixed effects due to the limited temporal scope (four cycles), to prioritise cross-country variation.

Benchmarking progress involved comparing T-scores and C-scores against annual targets, where countries failing to meet linearly projected milestones were classified as “not on track”. C-scores were treated as continuous variables, following AUDA-NEPAD’s standardised methodology, to analyse their marginal impact on hunger reduction. Key assumptions included linear relationships between variables and the absence of unobserved time-varying confounders. Summary of the method and approach is noted in the flow chart below (see Figure 2).

Limitations include the restricted temporal scope, constraining dynamic trend analysis, and the linear progress assumption, which may not fully capture nonlinear challenges like climate shocks.

By integrating CAADP’s hierarchical scoring system with panel regression techniques, this methodology provides a framework to evaluate how systemic factors—such as institutional capacity and financial access—influence hunger reduction, offering actionable insights for accelerating progress toward the Malabo Declaration’s 2025 targets.

3. Result

3.1. Trends in Regional Performance

The benchmarks for Commitment 3 progressively increased in each Biennial Review cycle, starting from 3.71 in 2017 and rising to 9.26 in 2023. Figure 3 presents the percentage of benchmarks met by the hunger-related T-scores across Africa for each Biennial Review cycle. These percentages declined over time, with T-scores of 1.79, 2.20, 2.71, and 2.90 corresponding to 48.25%, 43.65%, 42.88%, and 31.32% of benchmarks met in 2017, 2019, 2021, and 2023, respectively. Despite improvements in T-scores, the gap between actual performance and the benchmarks widened from 1.92 in 2017 to 6.36 in 2023. This reflects a growing divergence between actual performance and the targets set for hunger reduction.

3.2. Subregional T-Score Distribution and Variability

Figure 4 illustrates the distribution of T-scores across five African subregions, revealing variability in hunger reduction progress. Central Africa exhibits the lowest median T-score (approximately 1.3), with a broad interquartile range (1.0–1.6), suggesting significant disparities in performance within the region. However, East Africa reports the highest median T-score (approximately 3.0) and a narrower interquartile range (2.5–3.2), indicating more consistent performance across countries. Northern, Southern, and Western Africa display median T-scores ranging from 2.5 to 2.7, with comparatively smaller variability, reflecting a more homogeneous hunger reduction performance within these regions.

3.3. Rate of Improvement Across Biennial Review Cycles

The rate of improvement for T-scores was calculated across three Biennial Review intervals (2017–2019, 2019–2021, and 2021–2023), as shown in Figure 5. During the first interval (2017–2019), all subregions, except Central Africa (+10.87%), showed no improvement in their T-scores. In the second interval (2019–2021), West Africa exhibited the most significant increase (+85.16%). However, during the third interval (2021–2023), Central Africa experienced a sharp decline (−17.98%), while East Africa (+3.23%), Southern Africa (+14.34%), and West Africa (+5.92%) showed modest improvements.

3.4. Performance Across Hunger-Related Performance Categories (PCs)

Figure 6 presents the C-score-to-benchmark ratios for hunger-related Performance Categories (PCs) across regions and Biennial Review cycles. This ratio represents the proportion of the benchmark value achieved in a given year, calculated by dividing the C-score by the corresponding benchmark. A ratio of 1.0 indicates that the benchmark was met; values below 1.0 reflect underperformance, while values above 1.0 indicate that the benchmark was exceeded. To aid interpretation, these ratios are expressed as percentages in this section. The heatmap indicates substantial underperformance in several PCs across multiple regions. Specifically, West Africa met or exceeded the benchmark for PC 3.6 in 2023, surpassing the benchmark by 10%, while Central Africa and North Africa recorded significant underperformance in most categories. Central Africa achieved only 7% and 5% of the benchmarks for PC 3.4 in 2017 and 2019, respectively, and 7% for PC 3.3 in 2021. North Africa achieved 6% of the benchmark for PC 3.6 in 2023. Across all regions, PC 3.5 had the highest benchmark exceedance in 2017, with Africa meeting the benchmark collectively in all subregions.

3.5. Econometric Results

Full regression results are reported in

Table 1. Random Effects Panel Regression Results.

Performance Category (PC)	Coef. (β)	Std. Err	t-Stat	p-Value	95% CI Lower	95% CI Upper
PC 1.1—Country CAADP Process	0.002	0.031	0.075	0.940	−0.058	0.063
PC 1.2—CAADP Cooperation, Partnerships and Alliances	0.081	0.033	2.463	0.015	0.016	0.145
PC 1.3—CAADP based Policy and Institutional Review	0.069	0.042	1.638	0.104	−0.014	0.153
PC 2.1—Public Expenditure to Agriculture	0.027	0.042	0.646	0.520	−0.056	0.111
PC 2.2—Domestic Private Sector Investment in Agriculture	0.007	0.031	0.214	0.831	−0.055	0.068
PC 2.3—Foreign Private Sector Investment in Agriculture	0.059	0.029	2.020	0.045	0.001	0.117
PC 2.4—Access to Finance	0.062	0.039	1.574	0.118	−0.016	0.139
PC 4.1—Agricultural GDP and Poverty Reduction	0.132	0.050	2.645	0.009	0.033	0.231
PC 4.2—Inclusive PPPs for Commodity Value Chains	0.055	0.023	2.357	0.020	0.009	0.101
PC 4.3—Youth Jobs in Agriculture	0.019	0.024	0.794	0.429	−0.029	0.068
PC 4.4—Women Participation in Agribusiness	0.020	0.025	0.800	0.425	−0.029	0.069
PC 5.1—Intra-African Trade in Agriculture Commodities	0.008	0.032	0.243	0.808	−0.056	0.071
PC 5.2—Intra-African Trade Policies and Conditions	−0.067	0.032	−2.096	0.038	−0.131	−0.004
PC 6.1—Resilience to Climate-Related Risks and Shocks	0.043	0.027	1.599	0.112	−0.010	0.097
PC 6.2—Investment in Resilience Building	0.068	0.036	1.865	0.064	−0.004	0.140
PC 7.1—Country Evidence-Based Planning and M&E	0.038	0.019	1.938	0.055	−0.001	0.076
PC 7.2—Peer Review and Accountability	0.041	0.033	1.226	0.222	−0.025	0.106
PC 7.3—Biennial Agriculture Review Process	0.024	0.039	0.619	0.537	−0.053	0.101

Estimates from a Random Effects panel regression. Model selection was based on the Hausman test (p = 1.000). Sample size: 220 observations (55 countries across four biennial cycles). The dependent variable is the Commitment 3 T-score (0–10 scale). Independent variables are C-scores from non-hunger thematic areas, also scaled from 0 to 10.

.

The panel regression analysis investigates the relationship between performance in non-hunger thematic areas and progress in hunger reduction, as measured by Commitment 3 T-scores. The Hausman test produced a p-value of 1.000, indicating no significant difference between Fixed Effects (FE) and Random Effects (RE) coefficients, validating the use of the RE model. This suggests that unobserved country-specific factors (e.g., geographic or institutional traits) are uncorrelated with the predictors, allowing the RE model to efficiently estimate both time-varying and time-invariant variables. The model demonstrates strong explanatory power, capturing 94.3% of the total variation in T-scores (R² overall = 0.9433), with near-perfect between-country explanatory capacity (R² between = 0.9996) and substantial within-country explanatory power (R² within = 0.7559). The F-statistic of 36.376 (p < 0.001) confirms that the predictors collectively have a statistically significant impact on hunger reduction.

Five performance categories emerged as statistically significant at the 5% level. Agricultural GDP and Poverty Reduction (PC 4.1) showed the strongest positive association, underscoring the critical role of economic growth and poverty alleviation in food security. Multistakeholder Coordination (PC 1.2), Foreign Private Investment (PC 2.3), and Inclusive Public–Private Partnerships (PC 4.2) also demonstrated positive effects, highlighting the importance of collaborative governance, private-sector engagement, and cross-sector partnerships. Conversely, Trade Policies (PC 5.2) exhibited a negative coefficient, suggesting that trade liberalisation may inadvertently undermine food security by exposing smallholder farmers to market volatility or prioritising export crops over domestic needs. Investment in Resilience (PC 6.2) and Capacity for Planning and Monitoring (PC 7.1) were marginally significant at the 10% level, indicating potential secondary roles for climate adaptation and governance systems. The remaining predictors showed no statistically significant impact (p > 0.05), possibly due to indirect effects, heterogeneous country-level dynamics, or longer-term horizons required for measurable outcomes. Full regression results are reported in Table 1.

Our findings emphasise actionable priorities for policymakers, including targeted investments in agricultural productivity, strengthened multistakeholder coordination, and recalibrated trade policies to balance global integration with local food security. The model’s robustness reinforces the CAADP Biennial Review framework as a vital tool for tracking progress toward the Malabo Declaration’s 2025 targets.

4. Discussion

The Malabo Declaration is built on the principle that progress in one commitment area can reinforce outcomes in others. Commitment 3, which is focused on eliminating hunger by 2025, consolidates ambitious targets on productivity, inputs, irrigation, nutrition, food safety, social protection, and post-harvest loss reduction [16]. Based on the key information from available studies [10,11,16,17]. This study takes this understanding further and reflects on the trend analysis, which confirms that performance on Commitment 3 has remained below target across regions. While some countries have improved in selected non-hunger performance areas such as policy coordination, agricultural finance, trade, and resilience [11,13,14,15,16]; however, the extent to which these gains translate into hunger reduction remains unclear. Without empirical validation, it is uncertain whether the Malabo framework functions as an integrated system or as a collection of parallel targets. This study addresses that gap by assessing whether variation in country performance on non-hunger thematic areas (C-scores) is associated with outcomes under Commitment 3 (T-scores). Using a random effects panel regression model, the analysis identified five significant predictors PC 4.1 (agricultural GDP and poverty reduction), PC 1.2 (multistakeholder coordination), PC 2.3 (foreign private investment), PC 4.2 (inclusive public–private partnerships), and PC 5.2 (trade policies), as well as two marginally significant ones: PC 6.2 (investment in resilience) and PC 7.1 (Capacity for Planning and Monitoring). Performance categories under Commitment 3 (PCs 3.1 to 3.6) were excluded to maintain model independence. While these associations are significant, they are correlational, not causal. The sections that follow examine how each of these predictors may help explain cross-country variation in hunger reduction outcomes, drawing on both empirical evidence and contextual interpretations of their policy relevance.

4.1. Agricultural GDP and Poverty Reduction (PC 4.1)

PC 4.1 shows the strongest statistical association with Commitment 3 outcomes, making agricultural gross domestic product (GDP) and poverty reduction the most influential predictors of hunger reduction across countries from our analysis. PC 4.1 integrates five indicators. These include the growth rate of agricultural value added, agriculture’s contribution to national poverty reduction, reductions in poverty headcount at national and international poverty lines, and the reduction of the price gap between farmgate and wholesale markets. PC 4.1 yielded the most significant, most significant positive coefficient in the regression, which aligns with extensive evidence that agricultural growth continues to be one of the most effective drivers of poverty reduction in Sub-Saharan Africa [21,22,23,24]. Unlike redistribution-based approaches, which can be costly and unsustainable in mass poverty, agriculture-led growth offers a more durable and dignity-preserving pathway for income generation among rural populations [25]. This makes pro-poor agricultural growth a more feasible strategy for reducing hunger where labour absorption and asset access are possible. Dorosh and Thurlow [21] show that agriculture’s poverty–growth elasticity exceeds that of trade or manufacturing in countries such as Malawi, Mozambique, and Uganda, due to its labour intensity and direct engagement with rural populations. Cross-country analysis by Christiaensen and Martin [26] reinforces this view, showing that agriculture-led growth is particularly effective in low-income settings with shallow food markets and high underemployment, which are conditions prevalent across much of the continent.

Agricultural growth can be reduced by raising rural incomes, lowering food prices, and expanding food access [21,25,26,27]. In economies where agriculture remains the primary employer and income source for food-insecure households, its performance is closely tied to poverty dynamics [24]. Christiaensen and Martin [26], using sectoral GDP and labour decompositions, found that agricultural expansion generates the highest returns in poverty reduction due to its disproportionate employment of low-income groups. Dorosh & Thurlow [21] reached similar conclusions using economywide simulation models, showing that agriculture-led growth narrows consumption inequality and reduces the effects of real food prices, especially in settings where non-agricultural sectors benefit wealthier populations. Christiaensen and Martin [26] add that these effects are strongest when growth is concentrated in labour-intensive, market-linked, and nutritious crops. The strength of PC 4.1 in the regression thus reflects the size of agricultural growth, and its alignment with structural poverty conditions and rural food access constraints.

Beyond its direct effects on income and employment, agriculture also operates as a transmission channel for macroeconomic stability. Ardeni and Freebairn [28] show that agriculture absorbs and transmits both domestic and external shocks through its influence on input costs, consumer prices, and fiscal flows. This reframes PC 4.1 as a sectoral indicator and a proxy for broader macroeconomic dynamics such as inflation control, trade balance, and currency stability, which critically shape food affordability. Akpan et al. [29] in their study within the African context, reflected that agricultural performance is closely linked to inflation management, credit accessibility, and reserve adequacy. Rising agricultural GDP, they argue, can improve investor confidence, stabilise rural input markets, and reduce food price volatility. Noah et al. [30] add that while short-term gains in supply and affordability are evident, long-run impacts on food demand hinge on capital formation, infrastructure, and coherent macroeconomic policies. These findings collectively explain why consistent agricultural growth is associated with improved hunger outcomes in our analysis.

However, the hunger-reducing impact of agricultural growth is not automatic. It depends critically on the structural composition of that growth and its capacity to sustain inclusive rural linkages. Muba and Daudi [31], in their analysis of Tanzania, show that although the agriculture sector grew at 3.9% annually between 2006 and 2014, its share of GDP declined as other sectors expanded more rapidly. This form of structural transformation, while consistent with economic diversification, risks diluting the direct poverty and hunger reduction benefits of agricultural growth unless targeted investments preserve smallholder participation and rural employment. Berardi and Marzo [32] highlight a similar concern across African economies, demonstrating that agricultural GDP growth lacking labour absorption or regional inclusivity tends to yield weaker poverty impacts. In the context of PC 4.1, this suggests that countries scoring highly on sectoral growth may still underperform on hunger outcomes if the gains are concentrated in capital-intensive export subsectors with limited food system spillovers.

Evidence from African economies shows that agricultural value-added growth can trigger investment across related sectors, increase rural demand, and improve food security through diversified systems, as documented by Berardi and Marzo [32] and supported by multi-sectoral simulation studies such as those by Dorosh and Thurlow [21]. Ableeva et al. [33] explain that in structurally agrarian economies, stable sectoral growth can reduce import dependence and anchor market confidence. While these findings are not region-specific, they support the view that PC 4.1 may act as a transmission vector between agricultural strategies and food system outcomes.

4.2. Investment and Market Access for Smallholders (PC 2.3 and PC 4.2)

The performance of PC 2.3 (Foreign Private Sector Investment in Agriculture) and PC 4.2 (Inclusive Public–Private Partnerships for Commodity Value Chains) in the regression results reflects the growing policy emphasis on investment-led strategies within the Malabo Declaration. PC 2.3 captures the ratio of foreign direct investment (FDI) to agricultural value added, while PC 4.2 measures the number of value chains linked to smallholder farmers through established public–private partnerships (PPPs). Both categories are intended to connect capital, markets, and production that enhance productivity, income stability, and food access.

Several empirical studies support the association between foreign investment and improvements in agricultural output. Hanif & Nisa [34], using panel data from 15 developing countries, show that FDI and GDP per capita exhibit positive long-run associations with food output, though short-term effects remain unstable and context-dependent. Similarly, the NEPAD Eliminating Hunger in Africa report notes that countries with strong domestic investment platforms, such as Rwanda under CAADP, have attracted significant private capital, contributing to yield improvements and favourable food balances [35]. These findings suggest that foreign investment can reinforce hunger reduction when aligned with public infrastructure and agricultural policy coherence. Noah et al. [30] explain how capital accumulation, when directed toward physical infrastructure and technology, improves food system resilience and adaptive capacity. Nonetheless, the direction of causality remains ambiguous as countries with stronger food security systems may be more attractive to investors, particularly where agricultural governance is stable and regulatory incentives are predictable.

Concerns have been widely raised in the literature that foreign direct investment (FDI) in African agriculture is not inherently pro-poor. Tiba [20] and Simola et al. [19] argue that trade liberalisation under the African Continental Free Trade Area (AfCFTA) may encourage investment flows that favour large-scale, export-oriented agricultural production. This shift risks displacing smallholder farmers and undermining local food system sovereignty. Key dynamics of concern include land dispossession, the exclusion of producers who lack the capacity to operate at market scale, and the restriction of farmer autonomy through stringent intellectual property regimes [36]. This brings into sharper focus the role of public–private partnerships in shaping how market access and investment reach the rural people experiencing poverty.

PC 4.2 addresses precisely these institutional pathways. PPPs are widely promoted as mechanisms to overcome public capacity constraints by delivering infrastructure, inputs, and market access through collaborative arrangements. Obayelu [37] showed effects of such initiatives with Nigeria’s youth agribusiness schemes and Ghana’s cocoa and sorghum programmes, which aim to reduce transaction costs, expand production, and stabilise incomes through aggregation and purchase agreements. Amuda and Parveen [38], likewise, highlighted the catalytic role PPPs can play in food security by mobilising private finance in support of smallholder integration; however, there is potential for different outcomes of such initiatives across countries.

Comparative case studies offer insight into these variations. In Kenya’s Horticulture Development Programme (KHDP), PPPs linked producers to export markets via intermediary firms, but the model imposed high compliance costs, which contributed to the exit of smallholders from the value chain [39]. Rwanda’s Sustaining Partnerships to enhance Rural Enterprise and Agribusiness Development (SPREAD) initiative, by contrast, promoted cooperative-led processing and direct engagement with speciality buyers, leading to more equitable value distribution and sustained smallholder benefits [39]. Thorpe [40] attributes these differences to “procedural justice”, implying that partnerships succeed not simply by linking farmers to markets, but by embedding trust, transparency, and farmer voice in decision-making. This view was reinforced by Dorosh & Thurlow [21], who showed how value chains linked to local input demand, such as maize and tobacco, generated stronger poverty-reducing effects than those reliant on capital-intensive or import-based processing [41]. Furthermore, in contexts of declining public investment in agricultural research, PPPs can de-risk technology development and dissemination, provided that innovations are tailored to smallholder production systems, as seen in the Water Efficient Maize for Africa (WEMA) programme.

4.3. Coordination and Capacity for Planning and Implementation (PC 1.2 and PC 7.1)

PC 1.2 strong and PC 7.1 marginal associations with Commitment 3 outcomes highlight the importance of governance coordination and evidence-based policymaking in shaping the effectiveness of agricultural transformation and hunger reduction strategies. PC 1.2 measures the existence and functionality of multi-sectoral, multi-stakeholder coordination platforms, while PC 7.1 captures countries’ capacity to generate and use agricultural data via the Agricultural Statistical Capacity Index. However, neither indicator reflects a direct intervention; both function as enabling conditions for coherent and adaptive planning of agricultural policies [42,43,44].

Countries that perform strongly on PC 7.1 tend to have institutional capacity to analyse sectoral trends, allocate resources strategically, and monitor policy effectiveness through robust evaluation systems [42]. This capability is especially critical under conditions of food system volatility, where governments must respond rapidly to climate shocks, food price surges, or nutritional deficits [42,43,45,46]. The ability to adjust policies in real time rather than react after delays depends on quality data and embedded feedback mechanisms. As Fanzo et al. [47] argue, monitoring and evaluation frameworks are most impactful when integrated into continuous learning cycles within governance systems. In this context, the significance of PC 7.1 likely reflects a country’s broader institutional maturity and ability to plan and course-correct.

However, coordination capacity is equally crucial for converting data into coherent action. PC 1.2 captures this dimension by evaluating whether countries have established formal platforms that bring together ministries, agencies, civil society, and private sector actors to guide agricultural development. While such platforms exist across most African countries, their effectiveness varies widely. Evidence from Ethiopia’s Agriculture Development Partners Linkage Advisory Councils (ADPLAC) shows that state-led coordination bodies often remain embedded in hierarchical bureaucracies that limit farmer representation and reinforce top-down technology transfer models [48]. Similarly, Hermans et al. [49], through a social network analysis of multistakeholder platforms in Rwanda, Burundi, and the Democratic Republic of Congo, demonstrate that while these platforms may appear inclusive, they often exclude powerful ministries and private investors from meaningful engagement, leading to fragmented decision-making and weak alignment with national agricultural strategies.

A broader comparative review by IFAD [50] confirms these challenges. It distinguishes two prevailing models of CAADP-aligned coordination: institutionalised, policy-oriented platforms common in Francophone countries, and informal, business-facing clusters typical of Anglophone contexts. The former exhibit greater sustainability and potential for policy integration, particularly where national funding mechanisms exist, as in Senegal’s’ Fonds Interprofessionnel pour la Recherche et le Conseil Agricoles’ (FIRCA) or ‘Côte d’Ivoire’s Fonds National de Développement Agro-Sylvo-Pastoral’ (FNDASP), but still suffer from low stakeholder participation. The latter are more agile and responsive but often lack political anchoring. Both face core constraints, including limited trust among actors, poor cross-level coordination, and misalignment of priorities. These findings suggest that PC 1.2′s statistical significance may be capturing deeper institutional qualities such as actor diversity, decision-making structures, and coordination legitimacy rather than platform existence alone.

Hermans et al. [49] also pointed out that coordination effectiveness hinges on both connectivity and whether collaborative centrality aligns with institutional authority. As such, countries where ministries of planning or finance are detached from CAADP processes, even well-designed coordination platforms may fail to mobilise resources or shape national policy. Furthermore, donor-driven or project-bound platforms are unlikely to support long-term strategy development or scale innovation, as noted in the report from IFAD [50]. High PC 1.2 scores may offer a misleading impression of progress in such contexts, masking weak inter-ministerial alignment and superficial implementation beneath procedural formality.

4.4. Trade Integration and Food Systems Coherence (PC 5.2)

PC 5.2 evaluates the policy and institutional conditions supporting intra-African agricultural trade, measured through the Trade Facilitation Index and the Domestic Food Price Volatility Index. Its goal is to assess whether countries are creating harmonised, stable trade environments that improve market access and enhance regional food security. In this study, PC 5.2 emerged as the only statistically significant non-hunger predictor with a negative coefficient, indicating that stronger institutional scores in this domain are associated with weaker performance on hunger reduction (Commitment 3). This counterintuitive result questions prevailing assumptions that trade liberalisation and institutional trade reforms naturally promote food security objectives.

Empirical evidence from across the region provides several explanations for this pattern. Longo and Sekkat [51] and Ogunniyi et al. [52] show that intra-African trade remains severely constrained by infrastructure deficits, policy fragmentation, and overlapping regional memberships that generate conflicting tariffs and regulatory regimes. These issues weaken trade creation effects and constrain the structural interdependence necessary for food systems integration. Even where formal trade agreements exist, they often fail to reduce transaction costs or improve food access. Ogunniyi et al. [52], based on the panel data across African regions, revealed no significant long-run association between intra-African agricultural trade and hunger reduction, and a negative coefficient between agricultural exports and the hunger index, suggesting that increased trade flows do not necessarily translate into improved food security. Instead, trade often benefits well-positioned producers and sectors, bypassing the most food-insecure populations.

This disconnect is reinforced by studies highlighting the gap between trade policy design and implementation. Olney, [53] emphasises that customs red tape, weak enforcement, and poor administrative coordination frequently undermine well-intentioned reforms. Many countries report compliance with trade facilitation goals while failing to harmonise procedures at operational levels. Simola et al. [19] comment that AfCFTA’s projected benefits in agri-food trade are unlikely to be evenly distributed, as cereal output is expected to decline in several regions, and food prices may outpace incomes for vulnerable groups, undercutting potential gains in food security.

The sectoral composition of African economies also limits the benefits of intra-regional trade. As Olney [53] notes, countries with large agricultural sectors often engage less in intra-African trade due to similar production profiles and limited product differentiation. This structural homogeneity reduces trade incentives and undermines the potential for regional reallocation of food surpluses and deficits. Egbendewe et al. [54] offer an alternative framing, emphasising trade’s potential as a climate adaptation mechanism. Their bio-economic model demonstrates that intra-regional trade can buffer climate-induced yield variability by reallocating food from surplus to deficit areas. However, this stabilising role depends on coordinated trade regimes, supportive infrastructure, and data transparency conditions, often lacking across the region. Therefore, the negative coefficient for PC 5.2 in this study should be interpreted not as a failure of trade itself but as evidence of a deeper, more profound institutional mismatch. Policies aimed at expanding intra-African trade too often prioritise macroeconomic integration at the expense of food system coherence. The underperformance of trade reforms in reducing hunger reflects the absence of supportive infrastructure, insufficient safeguards for smallholder producers, and limited alignment with nutritional goals. As Tiba [20] cautions, liberalisation without equity can marginalise producers, erode food sovereignty, and expose domestic systems to volatility.

4.5. Resilience and Risk-Sharing Mechanisms (PC 6.2)

PC 6.2 measures whether countries have established dedicated budget lines for resilience-building initiatives aimed at mitigating risks from climate shocks, economic disruptions, and social vulnerability. Although this indicator was only marginally significant in the regression, its conceptual relevance remains strong. However, the binary design of PC 6.2 limits its explanatory power as it offers little insight into the scale, quality, or effectiveness of those investments. As such, the relationship observed in this study may reflect symbolic compliance rather than substantive integration of resilience into food systems or agricultural planning.

A central limitation of current resilience strategies lies in their temporal disconnect. Adger et al. [55] argue that interventions focused on immediate outputs, such as infrastructure rehabilitation or insurance payouts, often obscure deeper social–ecological fragilities. Without governance frameworks that enable long-term learning, cross-sector coordination, and iterative policy adjustment, resilience efforts risk becoming symbolic rather than transformative. It remained unclear whether PC 6.2 is equipped to capture these institutional dimensions necessary in a way that supports long-term food system transformation.

Sono et al. [56] also show that vulnerability-readiness gaps remain particularly severe in West and Central Africa. These deficits are linked to weak infrastructure, limited institutional coordination, and inadequate adaptive planning. The authors stress that resilience relies as much on multilevel governance, spatial targeting, and systems thinking as it does on funding. In this context, binary performance categories such as PC 6.2 are unlikely to reflect the full spectrum of resilience-building.

Despite these challenges, the rationale behind PC 6.2 remains sound. Resilience investments such as irrigation systems, drought-tolerant crop varieties, and early warning mechanisms can mitigate hunger risks by stabilising production and protecting vulnerable populations [57,58,59]. Such investments help reduce food price volatility and strengthen government capacity to deliver targeted assistance during shocks. However, the hunger-reducing potential of these interventions depends on their alignment with broader food security strategies and governance reforms.

5. Gaps, Needs, and Relevance of This Study

This study underscores the importance of enhancing the implementation quality and institutional orientation of Malabo-aligned strategies. The observed disconnect between high performance scores under the Malabo Declaration and the actual state of hunger in the region indicates that policy success cannot be inferred from formal compliance or structural presence alone. Greater emphasis is needed on aligning agricultural growth with rural poverty dynamics, ensuring that productivity gains benefit smallholders and food-insecure populations rather than being captured by capital-intensive subsectors. Investment and partnership models, particularly those involving public–private partnerships and foreign direct investment, require deliberate redesign to address structural exclusion and to promote equitable participation. Coordination platforms must evolve beyond formal existence to influence decision-making processes and resource flows in meaningful ways. Similarly, trade policy must move beyond liberalisation narratives to integrate explicit food security objectives, as market access alone has not delivered consistent gains in availability or affordability. Finally, resilience-building must be embedded in adaptive governance systems, with sustained institutional commitment replacing symbolic budget commitments. Without such shifts, the transformative aims of such declarations risk being undermined by superficial implementation and uneven outcomes.

While this study identifies clear empirical associations, several limitations must be acknowledged. First, the Commitment 3 T-score used as the dependent variable is a composite indicator that combines multiple aspects of hunger reduction, including productivity, nutrition, and post-harvest losses. This aggregation makes it to determine which specific dimensions of hunger most respond to improvements in non-hunger areas. Second, the independent variables are performance scores that reflect observed progress, not direct measures of specific policies or interventions. As such, the analysis cannot uncover the mechanisms through which change occurs or assess the design quality of individual strategies. These limitations constrain our study’s ability to evaluate effectiveness, but do not diminish the value of the findings in revealing consistent cross-category linkages that merit further investigation.

6. Conclusions

This study provides robust empirical evidence that while progress in select non-hunger thematic areas—agricultural growth (PC 4.1), foreign private investment (PC 2.3), inclusive public-private partnerships (PC 4.2), and multi-stakeholder coordination (PC 1.2)—is associated with improved hunger outcomes under the Malabo Declaration, these linkages are far from deterministic. Quantitative analysis of 55 African countries from 2017 to 2023 reveals a concerning trend: despite localised gains, Africa as a region is demonstrably off-track to meet the Declaration’s 2025 hunger eradication target.

Specifically, while T-scores for Commitment 3 have modestly improved over time (from 1.79 in 2017 to 2.90 in 2023), the gap between actual performance and the progressively increasing benchmarks has widened significantly (from 1.92 to 6.36). Subregional disparities further compound this challenge, with Central Africa consistently lagging behind other regions and key performance categories such as access to inputs and technologies (PC 3.1) and social protection (PC 3.4) exhibiting widespread underperformance across multiple regions and review cycles. These trends strongly indicate that the mere presence of institutions, investments, or coordination structures does not guarantee tangible progress on hunger reduction.

The analysis highlights a critical need to shift the focus from institutional existence and policy alignment toward a more nuanced understanding of how these mechanisms are designed, how they are structurally integrated into holistic development strategies, and how effectively they reach food-insecure populations. The negative coefficient associated with Trade Policies (PC 5.2) serves as a cautionary example, suggesting that unchecked liberalisation may inadvertently undermine local food systems and exacerbate vulnerability. Therefore, this study firmly challenges the assumption that policy intentions alone can drive results.

As the African Union prepares for the post-2025 agenda, policymakers must prioritise context-specific, strategically aligned, and effectively targeted interventions grounded in rigorous impact assessment. This entails a fundamental shift from measuring progress by institutional presence to measuring progress by demonstrable reductions in hunger and improved food security for the most vulnerable populations, ensuring that future strategies are both evidence-based and demonstrably effective. The future policies must go beyond mere performance categories (PCs) and evaluate regional disparities and context-specific needs.