Regional economic communities are a fundamental part of the African development framework, and the EAC is widely seen as a litmus paper for regional integration-driven development. The expected theoretical benefits are based on savings due to scale, competitive discipline, and investment creation (Schiff & Winters, 2003). Ejones, Agbola, and Mahmood (2021) find new empirical evidence for a statistically significant positive effect of EAC regional integration on per capita income growth, but the effect varies significantly across countries and sectors. Krantz (2024) finds that deepening regional backward linkages and participation in global value chains have increased since 2015, but the EAC heavily leans on extra-regional demand.

Osabohien et al. (2021), however, dampen this optimism by citing residual inhibitors of intra-EAC trade: non-tariff barriers, lack of trade facilitation, and infrastructure deficit. Lwesya (2022) further warns that without targeted effort to acquire domestic and regional backward and forward linkages and upgrade into regional and global value chains, EAC member states will only gain a share of low levels of the value chain. These reservations will also hold for the more conventional economic concerns surrounding intra-African economic integration, which draw on the structuralist views expressed in Hausmann, Hwang, and Rodrik (2006) that, without an industrial policy, openness can entrench comparative disadvantage. Ouma, Kimani, and Manyasa (2016) highlight a further agricultural issue that can constrain intra-EAC transactions: the low value of agricultural trade, but also the barriers imposed by a lack of harmonization of SPS standards and poor management of post-harvest logistics.

The link between trade openness and economic growth is a matter of empirical debate and remains hotly contested in development economics. Ibrahim and Abdulmalik (2023) show that trade openness plays a role in growth alongside governance, suggesting that institutional conditions for growth are key when liberalization is in progress in the EAC. Otieno Jabuya, Sule, and Ndwiga (2023) focus on trade openness in agriculture, concentrating on the impact of export diversification on GDP growth in the EAC; the study concludes: “what you export is likely to matter” (Hausmann, Hwang, & Rodrik, 2006).

On the other hand, Huchet, Mouël, and Vijil (2018) express concern that traditional indicators of openness are likely to overestimate the growth effect where the composition of imports is primarily capital or consumer goods that displace domestic production. Carrasco and Tovar-García (2021) complement this by positing that what counts is not total exports but their composition and diversification; economies that depend on a limited range of primary commodities have a weaker elasticity of growth to trade. These distinctions emerge in EAC-specific study foci: for example, while Raghutla (2020) and Zahongo (2016) find evidence of the trade-growth relationship in the context of developing and Sub-Saharan African economies in general, Yameogo and Omojolaibi (2021) signal that poverty reduces the positive effects of this relationship because increased disparity as a result of policy-induced liberalization, absent tradewise investment and redistribution strategies, opens up critical distributional arenas. Odhiambo (2021) also raises doubt about the unconditional reception of export-led growth in the EAC, attributing this to part of the mismatch between structural transformation and trade infrastructure development.

The growth effect of FDI is readily stated to be via capital accumulation, technology spillovers, and productivity growth (Borensztein, De Gregorio, & Lee, 1998). On the EAC, Zaman et al. (2021) show that the growth effect of FDI by itself becomes more significant if the integration of the region is considered, being, in fact, augmented by exports of information technology and capital accumulation. Banday, Murugan, and Maryam (2021) come to the same conclusion in a wider BRICS-plus context, with FDI and trade openness having a joint positive effect on growth—though within an institutional threshold.

Asiedu (2002) demonstrates that the determinants of FDI in Africa differ from those in other developing regions, with institutional quality a decisive locational factor a precondition distinct from, but complementary to, the growth mechanism identified by Borensztein, De Gregorio, and Lee (1998), in which FDI raises growth through capital accumulation, technology transfer and productivity spillovers conditional on absorptive capacity; this institutional precondition has been supported by Ajide and Raheem (2016), who found that the FDI-growth relationship in ECOWAS was strongest in the presence of strong legal and regulatory institutions. Akinlo and Okunlola (2021) incorporate weaker institutions into the developing country paradigm, concluding that past trends of simple trade openness only hold where the quality of institutions is sufficiently high. When domestic mobilization is hindered by poor institutions, the growth benefits of FDI are constrained, a finding which Belloumi and Alshehry (2018) distinguish from the growth effects of domestic capital flows and find to be underestimated for economically uncompetitive countries with insufficient local sources of specific technological capabilities. Sahoo and Dash (2022) bring specific industry heterogeneity into the picture, although their evidence concerns the heterogeneity of FDI impacts on exports across sectors rather than direct growth-spillover estimates.

The new institutional economics has conclusively shown that governance has a causal relationship with growth over the long run. Kaufmann, Kraay, and Mastruzzi (2011) set out the methodology of cross-country measurement of governance. Ndulu and O’Connell (1999) empirically relate Africa’s growth tragedy to failures in governance. Knack and Keefer (1995) find that measures of institutional quality (property rights and contract enforcement) explain a significant portion of cross-country growth differences more than those that are trade- and investment-related.

In the EAC, Ibrahim and Abdulmalik (2023) explicitly include governance as a moderation variable, with results demonstrating that institutional quality can actively moderate the growth effects of trade openness and FDI. Abdelbary (2018) makes a case for Egypt by highlighting that reform of governance is necessary for sustainable growth in the medium- to long-term, over and above stabilisation in the short term. Adzima and Baita (2019) find this to be true for sub-Saharan Africa as a whole, with governance indicators having a direct positive effect on GDP growth. Afolabi (2019) builds on this with a system generalised method of moments (GMM) dynamic panel approach, revealing that West African economic development is dependent on governance, a finding which can be safely applied to the EAC, sharing the same frazzled institutional spectrum. Fawaz, Mnif, and Popiashvili (2021) offer an assessment of this where they minimise heterogeneity by denying marginal growth in less developed, poorer, and fragile economies.

The other pillar of governance that is at the same time complementary to institutions is macroeconomic stability. Muinga et al. (2024) argue that twin macroeconomic fiscal deficits are a hindrance to EAC growth once they hold back private investment. Through a structural vector autoregression (S-VAR) model approach to monetary policy in the EAC, Olamide et al. (2022) show that macro external shocks-commodity price shocks and exchange rate shocks-transmission to domestic output seems to be quite swift, justifying the need for a credible countercyclical regime.

The expansion of digital financial services marks a fundamental change to the mobilization and use of capital in developing economies. Ngong, Thaddeus, and Onwumere (2024) offer the most direct evidence on the EAC, finding a growth-enhancing relationship between fintech adoption and development in the partner states. Dumor et al. (2024) examine the EAC as part of the Eastern African Belt and Road corridor and find that adoption of information and communication technology (ICT) increases trade and growth, driven by broadband access and regional integration.

Chinoda and Kwenda (2019) use panel cointegration for 30 African countries and find that mobile phone adoption, economic development and bank rivalry jointly influence financial inclusion across the continent and, hence, that fintech has multiple transmission channels. A review of financial innovation in Africa from the perspectives of development, competition and efficiency shows that technological leapfrogging can partially offset limited traditional banking opportunities (Beck, Senbet, & Simbanegavi, 2015). Exploring branchless banking models in Kenya, analysts find that financial innovations are correlated with bank performance and systemic credit intermediation (Chipeta & Muthinja, 2018).

Qamruzzaman and Wei (2019) provide evidence that financial innovation moderates the relationship between financial inclusion and the development of financial market infrastructure in South Asia by significantly reducing transaction costs and expanding the market for investable funds. These mechanisms matter significantly for the EAC, where formal financial depth is limited and mobile money utilization is among the highest globally (European Investment Bank, 2018; Financial Stability Board, 2017).

No modern growth decomposition can be silent about ecological and physical limits to production. Mutumba, Mubiinzi, and Amwonya (2024) report a unidirectional long-run causal relationship running from electricity consumption to economic growth (the “growth hypothesis”) in the EAC over 1990-2021, signifying that energy access constitutes a binding growth constraint. In fact, Ajakaiye and Ncube (2010) show that infrastructural gaps in energy, transport and water are the single most binding constraint to structural transformation in Africa.

A further complication concerns climate change: climate adaptation through adopting the water-energy-food-environment (WEFE) nexus for sustainable development in the EAC system indicates that the decade could end with devastating climate shocks reducing agricultural value-added (Mperejekumana et al., 2024). By exploring the connections between carbon emissions, economic governance, and growth, Otim and Mutumba (2025) observe that weak environmental governance intensifies the carbon intensity of output with serious growth costs. These results are in line with the international approach of reducing its impact, namely, growth at a lower rate (UNCTAD, 2022).

Sectoral studies lend greater layers of texture. The other studies included in this special issue are: Mwangi, Chen, and Njoroge (2020), who analyse agricultural imports and productivity in Sub-Saharan Africa and argue that capital-intensive agricultural imports could boost productivity but might also crowd out maize and other farming activity without concurrent policies to enhance rural development; Ghimire, Lin, and Zhuang (2021), who discuss Bangladesh’s experience with agriculture trade and highlight the balance-of-growth-environmental trade-off that the EAC will need to weigh as it expands its agricultural exports; Maina et al. (2023), who focus on housing and mortgage markets and show how housing investment increases the linkage between mortgage financing and growth in EAC member countries—an urbanisation-growth nexus that is insufficiently studied; Alananga and Mutasa (2022), who compare financial sector development in EAC with that of the Southern African Development Community (SADC) and find that proximity and regulatory convergence lead to stronger financial deepening-growth effects.

This study was completed as a systematic review in which every step of the review was performed using the PRISMA 2020 guidelines. The systematic procedure of this review was conducted with no internal or external tampering due to the use of an existing formulated protocol detailing in advance such key parameters as: the review question, criteria for selection, search strategy, and methods for analysis. The figures presented in this study were generated using for this study were produced using Python-based data visualization techniques. No ethics committee approval was required because this study did not result in the generation of new human or animal data and only used already published data. Nevertheless, adherence to generally accepted methods for citation, reference, and acknowledgement, and the prevention of plagiarism was essential and was implemented within this paper.

A methodical literature survey of relevant publications was carried out between January 2020 and December 2025 to represent the period during and after the COVID-19 pandemic, through five electronic databases and discovery platforms: Web of Science Core Collection, Scopus, EconLit, Google Scholar, and the Directory of Open Access Journals (DOAJ). The search process took place between 15 June and 30 June 2026, and was finalized during the final update, conducted on 15 September 2026, prior to submission of the manuscript on 30 September 2026. A search string was established by creating three search blocks connected by the Boolean operator AND: [economic growth (the dependent variable)], [geographical scope (EAC and related member/countries)], and [explanatory factors]. The full search string developed was: (“Economic Growth” OR GDP OR “Gross Domestic product” OR “Per Capita Income”) AND (“East African Community” OR “EAC” OR “Burundi” OR “Kenya” OR “Rwanda” OR “Tanzania” OR “Uganda” OR “South Sudan” OR “Democratic Republic of Congo” OR DRC) AND (“Regional Integration” OR “Trade Openness” OR “Foreign Direct Investment” OR “FDI” OR “Governance” OR Institutions OR “Financial Technology” OR “Fintech” OR “Electricity” OR “Energy” OR “Fiscal Deficit” OR “Monetary Policy” OR “Climate Change” OR “ICT”). Searching varied fields according to each database (e.g., Topic in Web of Science Core Collection, TITLE-ABS-KEY in Scopus, Anywhere in EconLit, the first 200 ranked records per query for each query string in Google Scholar, and All fields in DOAJ). Only studies in English language published in peer-reviewed journals (or working papers/conference papers if applicable) published between 2020 and 2025 were considered in the electronic databases. This research favored sensitivity over precision during the identification phase by trying to retrieve the largest number of possibly relevant publications. Once 26 duplicate records were removed, 118 distinct publications were identified for screening, as shown in Figure 1. To make the review as thorough as possible, the study conducted backward citation searching by scrutinizing the bibliography of all selected studies and relevant prior review articles. In addition, the research performed forward citation searching on Google Scholar for each selected study. Records found through these additional methods are recorded as “Citation searching” in the PRISMA flowchart (Figure 1).

To be included, studies were required to meet the following criteria: i) empirical peer-reviewed studies and closely related working papers; ii) be about EAC or one of its member states; iii) quantitative (using a statistical analysis approach such as OLS, panel data, time series) or qualitative; iv) utilize a robust method, showing clearly the relationship (cause and effect) between the chosen variable and GDP or economic growth with original data; and v) be related to a macroeconomic concept with respect to economic growth. Exclusion criteria applied as follows: non-empirical literature (editorials, commentaries, purely theoretical); if EAC countries are part of the sampled universe but represent a very minor group (<10%); without the disaggregated results; studies outside 2020-2025; unavailable working papers and duplicate papers.

Each retrieved paper was initially checked for title and abstract by the two reviewers and then assessed by the full text when it passed the former selection stage. Disagreements were resolved through discussion. Data extraction was conducted using an organized Excel data sheet in which bibliographic information was entered, as well as the design, time period of the study, indicators selected and their expected coefficients of relation with GDP/economic growth, the technique used (e.g., GMM, OLS, ARDL, VAR), direction and significance of effect(s), and limitation(s) stated by researcher(s). Inter-rater reliability coefficient was measured using Kappa, and the coefficient found was 0.87; this demonstrates a high degree of agreement.

To ensure a prioritization of evidence of high rigor in the synthesis of its core policy implications, the four weighted rating dimensions (Methodological Rigor (33.3%), Data Quality (25.0%), Analytical Depth (25.0%), and Policy Relevance (16.7%)) were scored for each included study using a 4-point Likert scale. At the corpus level, the studies scored highest across the four dimensions in the domains of Data Quality (mean = 3.5/4.0) and Policy Relevance (mean = 3.4/4.0), but exhibited greater heterogeneity across the measures of Methodological Rigor (mean = 3.2/4.0) and Analytical Depth (mean = 3.1/4.0). Individual weighted composite measures ranged from 2.85 to 3.75/4.0. Outlier studies (e.g., Ejones et al., 2021; Olamide et al., 2022) likely held more analytical weight, while findings from studies with lower scores were interpreted with explicit methodological caveats.

Each included study was scored independently by two reviewers on four dimensions using a 4-point Likert scale with the anchor rubric in Table 1: a score of 4 denotes a dimension fully met with transparent reporting; 3 denotes met with minor omissions; 2 denotes partially met with notable limitations; and 1 denotes not met or not reported. Discrepancies were resolved by discussion (inter-rater reliability reported in Section 3.4: Kappa = 0.87). The dimension weights (Methodological Rigor 33.3%; Data Quality 25.0%; Analytical Depth 25.0%; Policy Relevance 16.7%) were fixed a priori in the review protocol. Methodological Rigor carries the largest weight because the credibility of any causal claim about growth determinants rests primarily on research design, model specification, and diagnostic testing; Data Quality and Analytical Depth carry equal intermediate weights because reliable inference requires both sound measurement and robustness of estimation; and Policy Relevance carries the smallest weight because it conditions the usability, but not the validity, of a study’s findings. The weights sum to 100% and were combined into a weighted composite score (range 1 - 4) per study. Composite scores did not determine inclusion, which was governed solely by the eligibility criteria in Section 3.3; they shaped the synthesis in two ways: i) findings from the highest-scoring studies received greater interpretive weight in Sections 4.2 - 4.5 (e.g., Ejones et al., 2021; Olamide et al., 2022); and ii) findings from studies scoring below the corpus median were reported with explicit methodological caveats.

Table 1. Quality appraisal rubric (4-level anchors).

Note: A four-point scale was used to assess performance on each dimension: 1 = Not met, 2 = Partially met, 3 = Mostly met, 4 = Fully met. The weighted dimensions include methodological rigor (33.3%), data quality (25.0%), analysis (25.0%), and policy relevance (16.7%), which sum to 100%.

A formal meta-analysis was not feasible because of the diversity in research methodologies and the specifications of the macroeconomic factors across included studies. Thus, a narrative synthesis was constructed, grouping studies into themes reflecting the research questions, examining a frequency distribution table and a matrix of directional effects, and mapping research techniques to establish the best methodology for research to be conducted on the related factors of GDP/economic growth for these specific African countries.

This systematic search identified 118 records. After 26 duplicates were removed, 92 unique records were screened by title and abstract, and 22 studies were excluded (12 not EAC-related, 6 not empirical, 4 not within the date range). Seventy full-text articles were screened for eligibility, and 54 were excluded (28 not sufficiently focused on EAC, 12 not empirical, 8 outside the temporal period, 6 were insufficiently data-rich for synthesis), leaving a corpus of 16 empirical studies (13 studies focused exclusively on the EAC and 3 compared EAC member states with other SADC region economies). The PRISMA 2020 flow diagram is presented as Figure 1.

Table 2 summarizes bibliometric and methodological features of the 16 studies included in the review. The corpus of literature is somewhat limited in terms of a predominantly quantitative set of methods: panel data analysis (six); econometric time-series or ARDL models (five); structural VAR models (two); and multi-region input-output analysis (one). Two studies employ mixed or policy-analytical research designs. Study locations in the literature include, most prominently, Kenya, Tanzania and Uganda (seven), while Burundi, South Sudan and the Democratic Republic of Congo remain understudied. The largest theme addressed by the analysis is trade and regional integration (seven studies), followed by financial technology and governance (four) and energy and climate (three).

Table 2. Characteristics of included studies.

Note: WEFE = water-energy-food-environment; SADC = Southern African Development Community; WAEMU = West African Economic and Monetary Union.

Throughout Sections 4.2 - 4.5, each finding is identified by its dependent variable: the growth rate of real GDP, the level of real GDP, or real GDP per capita and, where reported by the original authors, by the estimated coefficient with its standard error or statistical significance. Specifically: i) Ejones, Agbola and Mahmood (2021) explain the growth rate of per capita income over 1988-2017 using feasible generalized least squares (FGLS) and panel-corrected standard error (PCSE) estimators; EAC regional trade agreement membership attracts a positive, statistically significant coefficient, with heterogeneous effects across partner states [exact coefficient and standard error to be quoted from the results table of the source paper]; ii) Otieno Jabuya, Sule, and Ndwiga (2023) estimate cross-sectionally augmented ARDL (CS-ARDL) models of real GDP growth in five EAC states over 2000-2021; agricultural trade openness carries a positive and statistically significant long-run coefficient [exact coefficient to be quoted from the results table of the source paper], and bootstrap panel Granger causality tests establish a unidirectional causal flow from agricultural trade openness to economic growth; iii) Ngong, Thaddeus, and Onwumere (2024) estimate a panel ARDL model of real GDP per capita over 1997-2019 and report statistically significant long-run coefficients of 5.87 for automated teller machines, 29.99 for debit card ownership and 5.25 for internet banking, a statistically insignificant effect for point-of-sale terminals, a negative long-run effect for mobile banking, and a negative and statistically significant error-correction term confirming long-run convergence; iv) Mutumba, Mubiinzi, and Amwonya (2024), applying panel DOLS, FMOLS and NARDL estimators over 1990-2021, establish long-run cointegration between electricity consumption and economic growth with a unidirectional causal flow running from electricity consumption to economic growth (the “growth hypothesis”); and v) Muinga, Gathiaka, and Osoro (2024), applying the pooled mean group (PMG) estimator over 2000-2021, find a positive and statistically significant (1% level) long-run association between fiscal deficits and growth at the regional level, alongside negative short-run effects in Burundi, Kenya and Rwanda. Where the original studies report only significance statements without full effect estimates, this is stated explicitly rather than inferred.

The empirical evidence analyzed here appears to strongly imply that regional integration, combined with trade openness, typically leads to a constructive, statistically significant expansion of EAC growth (Ejones et al., 2021; Osabohien et al., 2021). However, the evidence presented is not absolute: it implies that internal bloc backward linkages remain fragmented (Krantz, 2024), while sectoral literature hints at agricultural trade liberation potentially driving growth among largely agricultural economies (Otieno Jabuya et al., 2023; Ouma et al., 2016). Crucially, the interaction between trade openness and government efficiency in some contexts is positively and statistically significantly established (Ibrahim & Abdulmalik, 2023). Figure 2 illustrates the thematic significance.

FDI is always growth-enhancing in some manner; however, the channels are always found to be related to domestic financial depth and quality of governance (Ibrahim & Abdulmalik, 2023; Alananga & Mutasa, 2022). Likewise, the studies citing ICT and fintech have found positive spillovers on FDI in the presence of digital infrastructure and clear regulations (Ngong et al., 2024; Dumor et al., 2024). At the same time, the findings have indicated that FDI in sectors such as extractive sectors or enclave manufacturing-type FDI may result in higher measured GDP but may not necessarily generate productivity spillovers (Sahoo & Dash, 2022, who examine FDI impacts on exports rather than growth directly), which makes linkages to domestic small and medium firms crucial.

Among various macroeconomic shocks, fiscal deficits have been observed to hamper growth when financed through domestic borrowing that crowds out private credit (Muinga et al., 2024). External shocks weaken monetary policy transmission, whereas the shallowness of financial markets impairs the effectiveness of the policy (Olamide et al., 2022). Good governance enhances the effectiveness of both trade policies and investment policies (Otim & Mutumba, 2025). Together, these insights imply that macroeconomic management in the EAC can only be effectively carried out in the context of the institutional landscape-a finding akin to the new institutional economics literature (Kaufmann et al., 2011; Knack & Keefer, 1995).

Energy consumption (electricity) is found to Granger-cause growth unidirectionally, indicating that energy use is a binding growth constraint (Mutumba et al., 2024). Climate adaptation is emerging as a growth channel, with acceleration of environmental damage representing a long-term threat to agriculture (Mperejekumana et al., 2024). Investment in housing is identified as a notable growth channel, with mortgage market development stimulating construction industry users of cement and steel and household wealth impacts supporting household consumption expenditure and living conditions, as shown in Table 3.

Table 3. Thematic synthesis of determinants of economic growth in the EAC.

Note: NTBs = non-tariff barriers; SPS = sanitary and phytosanitary.

The range of methodological approaches in the 16 selected studies shows a substantial dependence on standard panel procedures. As clearly highlighted in Table 2, fixed-effects and random-effects panel models are the prevailing methods, complemented by time-series/ARDL specifications (five studies), structural VAR models (two studies), and multi-region input-output analysis (Krantz, 2024). There is no evident exploration of non-linear estimators that could potentially address regime-dependent behaviours and potential structural breaks, such as panel threshold regression, Markov-switching models, and dynamic ARDL simulation approaches. Similarly to cross-sectional dependence, which is of prime methodological importance given the high economic interconnectedness of the EAC bloc, results on it are available in less than one-third of the considered studies, as indicated in Figure 3.

The descriptive results of the quality appraisal based on Table 4 and Figure 4 demonstrate that our corpus ranks highest on data quality (mean 3.5/4.0) and relevance to policy questions (mean 3.4/4.0) and is most heterogeneous on the quality items about methodology quality and depth of analysis (mean 3.2 and 3.1, respectively). This is expected because very few diagnostic, robustness, and non-linear specifications were applied.

Table 4. Quality assessment framework and weighting criteria.

Note: Table presents the quality assessment framework and weighting criteria.

Empirical Limits and Methodological Directions

An important empirical constraint in this literature synthesis is the temporal gap between publishing periods and the periods of econometric estimation. This systematic review only selected empirical papers published from the start of the post-pandemic recovery era (2020-2025). Yet, only 1 of the 16 studies reviewed by this paper explored panel datasets ending after 2022 (Otim & Mutumba, 2025), with the rest being 2022 or earlier.

This confirms that while our review is current relative to the year 2020 and onwards, empirical growth modeling in the EAC, which should define policy for the new post-pandemic world, largely remains an assessment of the pre-2022 period.

Urgent empirical work is needed to populate growth studies with newer, post-2022 dataset panels.

Finally, very few papers used non-linear and spatial econometric estimation methods, potentially omitting thresholds and spill-overs (Pesaran, 2015; Ullah & Akhtar, 2018). Furthermore, the countries least captured in the growth regressions appear to be Burundi, South Sudan, and the DRC, which are likely to be the most volatile countries with high potential marginal returns to growth (see Figure 5).

As shown in Figure 6, the publication years of the included studies are from 2020 to 2025. Annual publications continued to rise over the years. The decline in the year 2025 does not indicate a fall in research activity, but a gap in publishing research as many manuscripts could be in the process of peer review.

The findings of this review depict a cohesive, though intricate, picture of growth in the EAC. Economic growth is not the product of any one element but of the convergence of regional integration, composition of trade, quality of institutions, diffusion of technology, and the extent of physical infrastructure. This is consistent with endogenous growth theory that supports long-run prosperity through convergence of innovation, human capital and institutional quality (Romer, 1990; Mankiw et al., 1992). It is also consistent with the new institutionalist perspective that institutional quality is not an isolated correlate of growth but a fundamental factor (Kaufmann et al., 2011; Knack & Keefer, 1995).

An important conclusion is that trade and investment benefits are conditional. Regional integration and trade openness will not necessarily lead to enhanced growth; their translation into growth benefits can depend on the nature of border administration, regulatory harmonization, and domestic absorptive capacity (Ibrahim & Abdulmalik, 2023; Lwesya, 2022). The conditionality can explain why the remarkable EAC trade liberalization programme, which included a customs union and common market, did not provide uniform growth dividends to the partner states—some benefited more with better institutions and diversified production structures than those that are single-commodity-dependent.

Also, like FDI’s contribution to growth, the effect is non-automatic. The literature reaffirms the foundational work of Asiedu (2002) that the FDI-growth relationship is conditioned by institutional and infrastructural gaps in Africa, since Asiedu (2002) demonstrates that FDI inflows themselves depend on institutional quality. FDI targeting extraction or enclave manufacturing could be associated with an increase in measured GDP but not necessarily in related productivity spillovers, although the study notes that Sahoo and Dash (2022) provide evidence on the heterogeneity of FDI impacts on exports rather than direct evidence on growth spillovers. For the EAC, this implies a focus in investment promotion on technology spillovers, localization requirements, and linkages with small and medium-sized firms.

Finally, the rise of financial technology as a mediating pathway to development is a major thematic development in recent literature. Ngong et al. (2024) and Dumor et al. (2024) show that ICT adoption and digital financial services can in part overcome the shortcomings of traditional banking infrastructure to speed up capital accumulation and facilitate trade. Nonetheless, the fintech-growth channel is conditional on access to electricity, digital competence and regulatory frameworks that support innovation and consumer protection-a warning that technological leapfrogging cannot replace complements to infrastructure (Beck et al., 2015; Chinoda & Kwenda, 2019).

Although there is a clear general trend to expect a favourable impact of integration and trade on growth, surprising exceptions appear. Whereas Ejones et al. (2021) and Otieno Jabuya et al. (2023) find very strong positive trade elasticities, Huchet et al. (2018) and Odhiambo (2021) warn that indications of export-oriented competitiveness revealed by openness must be interpreted with care as potential indications of deindustrialization if import penetration exceeds export capacity. The differences could be the result of model specification, differences in time frames, and how a structural break is modeled. The fact that the literature on the EAC does not offer any non-linear estimators impairs the ability to reconcile the two views, as no estimate can be made to identify an institutional/infrastructural “floor” from which openness may foster growth.

The second tension arises from fiscal policy. While Muinga et al. (2024) show that fiscal deficits negatively impact growth, the Keynesian counter-cyclical approach implies that deficit-financed infrastructure investment could increase the economy’s potential if managed effectively. The papers reviewed do not sufficiently address this tension, partly because most studies rely on linear panel and time-series specifications (11 of 16; Figure 3) that are unable to differentiate between productive and unproductive government spending.

Third, both geography and theme represent glaring lacunae in the literature. The three most fragile and potentially highest marginal return economies in EAC are Burundi, South Sudan and the Democratic Republic of Congo, yet they remain under-represented in the empirical corpus. The informal economy-also the best-known route for employment for the majority of EAC workers-has been almost ignored by the growth regressions covered herein. Similarly, there are few high-frequency post-pandemic studies on the nexus between climate and growth, despite some momentum (Otim & Mutumba, 2025; Mperejekumana et al., 2024), which miss the level of granularity with country-specific modelling to drive adaptation policy.

The review surfaces four policy priorities. First, market liberalization must be ushered in and complemented by deepening the institutional architecture. A concerted push for reforms that enhance contract enforcement, mitigate corruption, and unify cross-border market protocols will raise the spillover benefits of regional integration and foreign direct investment (Abdelbary, 2018; Adzima & Baita, 2019; Kaufmann et al., 2011).

Second, macroeconomic stability must be preserved with credible fiscal rules and autonomous monetary systems that avoid deficit crowding-out and provide resilience to external shocks (Muinga et al., 2024; Olamide et al., 2022).

Third, a concerted industrial and agrarian strategic policy must aim at export-composition diversion and value chain upgrading of traditional export commodities (Carrasco & Tovar-García, 2021; Lwesya, 2022). Fourth, a major focus on digital and energy infrastructure as a growth anchor rather than a myopic residual should be a major priority: universal access to electricity and broadband, as well as a regulatory sandbox for mobile-payment and metaverse development, should be priorities (Mutumba et al., 2024; Ngong et al., 2024).

The limitations of this review include the following. English only: Non-English relevant research was excluded, such as Francophone or Lusophone research regarding EACs. Outcomes: Due to bias towards published research, non-null or policy-supportive findings, there was publication bias towards statistically significant findings. Methods: The variation in specifications was too great to Meta-analyze effect sizes. The time frame of 2020-2025: While successful in capturing the post-COVID pandemic period, this excludes valuable antecedent research (before 2020) that frames and contextualizes future EAC development literature. These limitations were mitigated by intensive multiple database searches, manually screening references, and including not only country-specific EAC literature but also related comparative EAC regional studies.