Purpose

This study investigates the linkage between structural transformation and inclusive growth in sub-Saharan Africa (SSA) by decoupling productive transformation, reflected in sectoral value-added and social transformation, reflected in cross-sector employment shifts. It asks whether changes in agriculture, manufacturing, industry and services value-added and employment shares transmit differently into inclusive growth.

Design/methodology/approach

The study constructs composite indices for inclusive growth and structural transformation to capture the multidimensional nature of these concepts. It uses annual unbalanced panel data for 32 SSA countries over the period 2000–2024. The empirical strategy applies the Bias-Corrected Method of Moments estimator as the main technique, supported by pooled ordinary least squares, between-country regression, common correlated effects, median quantile regression with year fixed effects and one-year lagged regressor model with year fixed effects estimation for robustness, with specifications adjusted to limit multicollinearity among sectoral shares.

Findings

The results establish a positive relationship between structural transformation and inclusive growth in SSA, suggesting that broad-based sectoral reallocation supports shared prosperity. The twin-channel evidence indicates that employment movement into services is the most consistent driver of inclusiveness, whereas manufacturing value-added has no statistically significant independent effect. The findings imply that the quality and employment-absorbing capacity of transformation matter more than the expansion of sectoral output shares alone.

Research limitations/implications

In the process, one potential limitation is that of availability and quality of sectoral employment data and data on value added across SSA countries at high frequency with consistent classification in terms of the sectors which might affect how granularly the analysis can be carried out. Furthermore, although the composite indices of inclusive growth and structural transformation were developed using state-of-the-art techniques, they might not record all the dimensions and complexities of these multifaceted phenomena. Next steps could involve studying alternative methods of analysis (e.g. panel threshold models or quantile regressions) in order to draw further conclusions with respect to the non-linear relationship between the process of structural change and inclusive growth. Second, taking a finer disaggregated perspective of the services sector in terms of high-productivity modern and low-productivity traditional services could offer a richer insight into inclusivity determinants. Future research might also replicate this dual-channel conceptualisation over more developing countries or on specific sub-regions in Africa to ascertain the generalisability of the results and compare contextual differences in the Structural Transformation-Inclusive Growth (ST-IG) relationship.

Practical implications

Policymakers should prioritise structural transformation strategies that emphasise employment creation rather than output expansion alone. Strengthening high-absorbing service sectors, improving service trade integration and enhancing productivity in agriculture are critical for inclusive outcomes. Industrial policies should focus on employment intensity and linkages rather than manufacturing expansion per se. Coordinated labour market, trade and human capital policies are essential to ensure transformation translates into shared prosperity.

Social implications

The findings highlight the importance of labour-absorbing structural change for reducing poverty and inequality in SSA. Shifting workers from low-productivity agriculture into productive service activities can improve livelihoods and social inclusion. Failure to align transformation with employment generation risks reinforcing exclusion. Inclusive structural change is therefore central to achieving decent work, social mobility and equitable development outcomes in the region.

Originality/value

This study contributes by jointly modelling sectoral value-added and employment shifts within a single SSA-focused inclusive-growth framework. By distinguishing the productive and social dimensions of structural transformation, it clarifies why sectoral upgrading may be inclusive only when output gains are matched by productive employment opportunities. The findings offer sector-specific policy guidance for structural transformation strategies aligned with Africa’s Agenda 2063.

The global agenda on development has gradually evolved from narrowly concentrating on growth to a broader focus of including the quality and sustenance of growth, underpinned in recent key development reports (The World Bank, 2022; UNDP, 2022). This paradigm underlies the United Nations Sustainable Development Goals (SDGs), notably SDG 8 to promote sustained, inclusive and sustainable economic growth, full and productive employment and decent work for all. Procuring these objectives, notably in low-income regions, is intrinsically bound to the process of structural transformation, referred to as the movement of economic resources out from agricultural and other low-productivity sectors into higher productivity activities such as industry or advanced services (McMillan and Rodrik, 2011; Herrendorf et al., 2013).

In principle, the process of structural transformation (including changes in value-added and employment shares across sectors) should generate growth which is inclusive in the sense that it absorbs workers out of low-productive activities and into higher productivity activities while also raising national average incomes (Lewis, 1954). Yet, a development paradox endures in Sub-Saharan Africa (SSA), such that despite modest economic growth during the past 20 years, the region is confronted with high poverty, unemployment and inequality rates, ranking second globally (AfDB, 2015; World Bank, 2022). It simply means that the structure of growth in SSA has largely been non-inclusive, telegraphing ominous questions around the character of its structural transformation.

The widening gap has increasingly been ascribed to the nature of structural transformation in SSA. Contrary to the development trajectories of more advanced economies, SSA has been characterised by a “premature deindustrialisation”, with an over-representation of low-productivity services rather than manufacturing (Rodrik, 2016; Cadot et al., 2016). As a result, even as employment has gradually agglomerated out of agriculture, the absorption has been predominantly into low-productivity sectors, making in turn the change “growth-reducing” (Rodrik, 2016). The suggestion is that a reallocation of labour is not enough, and the direction of shifts and sectoral sources of value addition are crucial inclusions for inclusivity.

The available empirical literature has placed a sound basis on the stimulating effects of structural transformation on growth. Unfortunately, it has not yet specifically disentangled the separate means through which transformation affects inclusive growth. There is not much in the literature which had explored structural transformation as a heterogeneous process by investigating value-added contributions (a proxy for productivity and output composition) and labour shifts (which can be used to reflect income distribution and livelihood opportunities) separately. This is a substantial blind spot as theory suggests that value-added growth in capital-intensive sectors (such as extractives) might even exacerbate inequality and that it could promote inclusivity when produced by labour-intensive sectors (for instance, manufacturing) (Baymul and Sen, 2019; Loayza and Raddatz, 2010). Likewise, the speed and form of employment shifts between sectors determine how the spoils of growth are distributed.

Although some studies like Magwedere and Marozva (2024), Atangana et al. (2025) and others have started investigating the inequality and structural transformation linkages however, their approach is narrow and confined to specific sectors or combined variables therein. As far as the authors’ limited knowledge goes, no careful study has simultaneously modelled these two dual channels of sectoral value-added and cross-sectoral employment shifts in an integrated manner to account for inclusive growth in SSA. This gap is increasingly relevant for regional policy instruments, such as the African Unions Agenda 2063 which calls to “Industrialise Africa” and to “Improve the Quality of Life,” but does not have evidence at sectoral level on which pathways are most effective in driving shared prosperity.

Hence, this study aims to contribute to literature by empirically examining how sectoral value-added contributions and employment shifts spur inclusive growth in SSA. The study is anchored on the following core research questions: (1) What are the general implications of structural transformation to inclusive growth in SSA? (2) How does value addition in agriculture, industry, manufacturing and services matter for inclusive growth differently? (3) How do changes in employment shares in agriculture, industry and services affect inclusive growth?

In this context, this study has three major contributions to literature:

  1. Dual-channel framework: The study goes beyond the aggregated computations of structural transformation by introducing a dual-channel analytical framework which discusses in an explicit manner, the contribution of sectoral value-added (productivity channel) from cross-sectoral reallocation of employment (distribution channel) to inclusive growth. It therefore offers a more fine-grained insight into how transformation and inclusivity are related.

  2. Extensive empirical analysis: We use robust panel econometric methods such as the Bias-Corrected Method of Moments (BC-MM) estimator proposed by Breitung et al. (2022). This estimator is highly appropriate here for our dynamic panel model with relatively short time dimension (T) and persistent variables, such as sectoral value-added. The BC-MM estimator automatically accounts for the finite-sample bias problem that is well-documented to afflict normal difference and system Generalized Methods of Moments (GMM) estimators in such environments, thereby delivering less-biased coefficient estimates overall for our analysis of both structural transformation and inclusive growth in SSA.

  3. Sector-specific policy guidance: The results present clear evidence-based policy guidance to policymakers regarding the differential effects of different sectoral pathways. What the results clearly show is that the agriculture, manufacturing, industry and service sectors are powerful engines of inclusive growth in terms of both value added and employment.

The rest of the study is organised as follows. In Section 2, the related literature is reviewed, and in Section 3, the data and methodology are described. Section 4 provides empirical results and discussion, followed by policy implications.

The structural transformation process, or the transition of an economy from primary to manufacturing, industry and services-based activities, is generally believed to be a fundamental aspect in economic development (McMillan and Rodrik, 2011). This change in resource allocation, as evidenced by changing sectoral value-added and employment, is essential to enable the productivity enhancements required for continued advances (Lin, 2011). For the region of SSA, where rapid expansion has not always been translated into broad-based welfare gains, it is important to understand this relationship between the pattern of structural change (adopted innovation) and the prospect for an inclusive growth process (Rodrik, 2016; McMillan et al., 2014).

The literature on structural transformation provides the theoretical basis for examining sectoral dynamics in SSA. Lewis (1954) and Kuznets (1955) remain the classical references because Lewis explains development as the transfer of surplus labour from traditional low-productivity agriculture to modern higher-productivity industry, while Kuznets links development to changes in inequality as resources shift towards modern sectors.

For this study, structural transformation is conceptualised as a dual but connected process, consistent with Lewis (1954), Kuznets (1955), Herrendorf et al. (2013) and McMillan et al. (2014). The first dimension is productive transformation, which refers to the reallocation of output and productivity across agriculture, manufacturing, industry and services (Chenery and Taylor, 1968; Timmer and de Vries, 2008; McMillan and Rodrik, 2011). The second is social transformation, which concerns the movement of labour, livelihoods and earnings opportunities across sectors (Gollin et al., 2016). This separation is necessary because a sector can expand output without absorbing labour, while another can absorb workers without generating high productivity or decent earnings (Rodrik, 2016; Diao et al., 2019). Therefore, the study treats sectoral value added as the productivity channel and sectoral employment shares as the distribution channel.

On the supply side, transformation operates through productivity, production linkages, capital deepening and technological upgrading (Kaldor, 1966; Hirschman, 1958; Ngai and Pissarides, 2007). On the demand side, it operates through income growth, labour absorption, consumption patterns and access to services (Kongsamut et al., 2001; Buera and Kaboski, 2012).

Its main critique is that the classical framework does not automatically explain inclusive growth, because movement out of agriculture may be growth-reducing where labour enters informal low-productivity services (McMillan and Rodrik, 2011). Another weakness is that early theories tend to assume a fairly linear movement from agriculture to manufacturing and then services, while SSA often experiences premature deindustrialisation, weak formal job creation and expansion of low-productivity services. The theory is nevertheless relevant because it justifies testing whether value-added expansion and employment reallocation move together, whether agriculture, manufacturing, industry or services dominate the productivity channel, and whether one channel explains inclusive growth more strongly in SSA. It also appropriately allows the paper to clarify the bidirectional relationship in which inclusive growth can strengthen human capital, domestic demand and fiscal capacity, thereby supporting later transformation.

Discussion of the theoretical links between structural transformation and inclusive growth has motivated a growing body of empirical work, although direct SSA evidence remains limited and findings are mixed. Global evidence is sensitive to measurement choices, transformation proxies and econometric techniques (Comin et al., 2021).

Trofimov (2023), using panel Ordinary Least Square (OLS) and Vector Autoregression (VAR) for 111 economies from 1971 to 2018, suggests that persistence in agriculture and excessive services expansion constrain Gross Domestic Product (GDP) growth, while industrialisation and faster structural transformation are growth enhancing; however, the wide sample risks over-generalisation for SSA. Similarly, Stawska and Jabłonska (2022) measure inclusive growth in EU-27 using a weighted correlation approach, but the method is subjective and largely descriptive. Rashid and Intartaglia (2017), using two-step system GMM, suggest that financial development reduces absolute poverty but not necessarily relative poverty, reinforcing the view that aggregate development does not automatically become inclusive. Dabús and Delbianco (2021) show that industry is generally favourable while some services may be harmful to growth while Basu and Nag (2020) connect sectoral inter-linkages and employment to fiscal and external conditions. These studies therefore justify separating output composition from employment channels, but they still leave unresolved whether structural transformation improves inclusive growth through productivity, labour absorption or both.

Furthermore, African studies reveal stronger contradictions around sectoral value added and employment. Magwedere and Marozva (2024), using system GMM for low- and middle-income African countries, found that mining and construction reduced inequality while agriculture and manufacturing unexpectedly increased it, whereas Mamman and Sohag (2023), using quantile regression, found that movement from agriculture to services reduced poverty but increased inequality and that manufacturing had no significant effect. These tensions show that services may absorb labour without necessarily reducing inequality and that manufacturing is not always inclusive in African conditions. Asongu et al. (2018), using system GMM for 49 SSA countries, found that remittances and information and communications technologies (ICT) reduce some constraints to doing business but increase other procedural burdens, while Ssozi et al. (2019) found that agricultural development aid improves productivity but may substitute food crops with industrial crops. Thus, agriculture matters when value addition is productivity-enhancing and institutionally supported, not merely because its GDP share rises. The evidence also indicates that institutions, economic freedom and digital connectivity may determine whether sectoral upgrading becomes socially broad-based or remains confined to narrow productive enclaves.

In addition, Morsya et al. (2023), using Demographic and Health Surveys (DHS) microdata for 37 African countries, found that higher agricultural employment shares increase between-country inequality, with no clear evidence for industry or services. Kagochi (2019) shows that inflation constrains financial sector development in SSA, while trade openness and good governance improve it. Asongu and Odhiambo (2019) similarly suggest that financial access supports insurance development through governance channels. Mateko (2025), focussing on the Maghreb, found that agriculture increases poverty in the short run, while services and manufacturing reduce poverty in the short and long run, but the Maghreb differs from SSA in infrastructure and institutional conditions. Gamette et al. (2025) add that remittances expand rural and urban electricity access in SSA, although governance shapes the effect. Enongene (2024) found that value added in agriculture, manufacturing, industry and services is pro-poor in SSA, with agriculture most effective, supporting Busse et al. (2019), who found that African structural change has occurred but contributed unevenly because labour remains concentrated in agriculture.

Conversely, Atangana et al. (2025) show that inequality predicts structural transformation and labour productivity, while Marfatia (2023) and Yerrabati (2024) suggest that sectoral performance, growth expectations, remittances and vulnerable employment interact dynamically. At the same time, the regional distinction matters because SSA differs from North Africa, Asia and Europe in terms of labour absorption, infrastructure access, informality and the depth of manufacturing. Studies on Europe or the Maghreb often emphasise formal institutions, services and manufacturing, while SSA studies repeatedly point to agriculture, remittances, infrastructure and governance. This contrast means that comparative evidence is useful but cannot replace SSA-specific estimation. It also explains why variables with weak effects in the study, especially manufacturing value added and aggregate employment shares, may be weak not because they are irrelevant, but because the sectors themselves remain structurally shallow or highly informal in the SSA sample. Therefore, the empirical debate is not whether structural transformation matters but which component of transformation is sufficiently productive and distributive to become inclusive.

Methodologically, the studies also differ in important ways. Some use panel OLS, VAR, Autoregressive Distributed Lag (ARDL) or descriptive decomposition, while others use system GMM, quantile regression and instrumental-variable approaches. This matters because structural transformation and inclusive growth are likely to be persistent and jointly determined, so static approaches may overstate or understate the size of sectoral effects. The literature also differs in its outcome variable: some papers study GDP growth, others poverty, inequality, financial development, vulnerable employment or infrastructure access. These outcomes are related to inclusive growth but do not fully capture the multidimensional index used in this study. For SSA specifically, the strongest unresolved issue is whether agriculture, manufacturing, industry or services matter because of output growth, employment absorption or both.

Overall, this literature leaves a gap because it rarely estimates sectoral value-added shares and employment shares jointly within a dynamic SSA inclusive-growth framework. This study responds by treating productive transformation and social transformation as related but distinct channels, thereby offering evidence on whether SSA’s inclusive-growth gains arise from sectoral productivity, labour reallocation or their interaction over time across diverse SSA economies, income levels, institutional settings and sectoral structures comparatively.

The analysis is based on secondary annual unbalanced panel data for 32 Sub-Saharan African countries (list of countries presented in the Online Appendix as B1) over the period 2000 to 2024, derived from the World Development Indicators (WDI) and other databases such as UNCTAD (United Nations Conference on Trade and Development). The study window is selected because annual data are available for the variables under investigation, and the choice of countries relies on reliable observations for the core variables over time to reduce information-induced bias and reinforce econometric robustness. Data description and sources are shown in Table 1. The variables reflect the transition of an economy from agriculture to manufacturing, industry and services, as well as whether growth translates into distributional and welfare improvement. The study discusses some of the key variables used in this study below.

Table 1

Variables, definition and data sources

VariableDescriptionsSource
Inclusive growthIndex of inclusive growthComputed based on Mlachila et al. (2016) 
Structural transformationIndex of structural transformationUNCTAD
Agriculture VAAgriculture, forestry and fishing, value added (% of GDP)WDI
Employment in agricultureEmployment in agriculture (% of total employment)WDI
Manufacturing VAManufacturing, value added (% of GDP)WDI
Employment in manufacturingEmployment in manufacturing (% of total employment)WDI
Services VAServices, value added (% of GDP)WDI
Employment in servicesEmployment in services (% of total employment)WDI
Industry VAIndustry (including construction), value added (% of GDP)WDI
Employment in industryEmployment in industry (% of total employment)WDI
InstitutionsIndex of institutional qualityUNCTAD
Trade in servicesTrade in services (% of GDP)WDI
InflationInflation consumer prices (annual %)WDI
Gross fixed capital formationGross fixed capital formation (% of GDP)WDI
Human capitalIndex of human capital (health, education, skills)UNCTAD
Source(s): Authors’ compilation

Inclusive growth is the dependent variable and is treated as a composite index rather than a single income proxy. In line with Mlachila et al. (2016), it captures economic fundamentals and social fundamentals, where the economic dimension reflects stable and welfare-enhancing growth, while the social dimension reflects broader opportunities and improved living conditions. Principal component analysis (PCA) is used to combine related dimensions into one score, with the retained factor axis representing the largest common variation and higher normalised values indicating stronger inclusive growth. This choice is important because GDP per capita growth alone cannot capture whether the benefits of growth are stable, sustainable and transmitted to welfare.

For structural transformation which is our explanatory variable, the study distinguishes between productive and social sub-indices before constructing the overall measure. Productive transformation is based on agriculture, manufacturing, industry and services value-added shares, and the first productive-transformation factor explains 45.6% of common variance, while the second factor raises cumulative explained variance to 79.9%. This implies that productive transformation is multidimensional and cannot be fully represented by one sector. Social transformation is based on employment shares in agriculture, industry and services, and its first factor explains about 87.4% of common variance, suggesting that employment reallocation is summarised by one dominant axis. The overall structural transformation measure can therefore be treated as the externally sourced structural transformation index used in the estimations or, where Principal Component Analysis (PCA) is used, as an index combining productive and social sub-indices. When the two sub-indices are combined, the first overall factor explains approximately 82.7% of their common variance.

The core regressors include agriculture value added, manufacturing value added, services value added, industry value added, employment in agriculture, employment in services and employment in industry. Control variables comprise institutional quality, trade in services, inflation, gross fixed capital formation and human capital. These variables are selected because they capture policy conditions, openness, macroeconomic stability, investment and capability formation that can shape the transformation-inclusive growth relationship. The use of both value-added shares and employment shares responds to observations that structural transformation contains supply-side and social dimensions. Value-added shares capture productive transformation, because they show how output and productivity are distributed across sectors. Employment shares capture social transformation, because they show where labour, livelihoods and earning opportunities are located. The data strategy therefore links measurement directly to the paper’s novelty.

The modelling strategy proceeds from diagnostics to benchmark and dynamic estimators. Because Pesaran cross-sectional dependence tests do not reject cross-sectional independence for the variables used, first-generation panel unit-root procedures are appropriate. Fisher-type Augmented Dickey-Fuller (ADF) and Fisher-type Phillips–Perron (PP) tests are used because they allow heterogeneous autoregressive parameters across countries and are suitable for unbalanced panels (Maddala and Wu, 1999; Choi, 2001). The Fisher-ADF test addresses serial correlation through lag augmentation, while the Fisher-PP test provides a non-parametric correction that is flexible under serial correlation and heteroscedasticity. Together, they determine whether the variables are stationary in levels or first differences. Cointegration is then tested because the variables show a mixture of stationary and first-difference stationary behaviour, and the study needs to establish whether inclusive growth and its structural drivers share a stable long-run relationship. This sequencing is necessary because unsuitable stationarity assumptions can produce misleading regression results, while evidence of cointegration justifies dynamic modelling of long-run relationships among variables that may adjust after short-run deviations.

To address endogeneity, the empirical strategy moves from descriptive and benchmark models to estimators that handle dynamic bias, unobserved heterogeneity and feedback between inclusive growth and structural transformation. Pooled Ordinary Least Square (OLS) and between-country regressions are retained only as benchmark models. Pooled OLS treats all observations as one dataset and is estimated with robust errors, but it does not control country-specific unobserved effects (Wooldridge, 2010; Baltagi, 2008). The estimation strategy uses the robust standard errors proposed by Driscoll and Kraay (1998) to control for heteroscedasticity. The between estimator exploits cross-country averages and is useful where cross-sectional variation dominates, although it cannot address dynamic persistence or simultaneity. Therefore, the main inference is based on the BC-MM, introduced by Breitung et al. (2022), which includes the lagged dependent variable to model persistence in inclusive growth and corrects dynamic panel bias. Reverse causality may arise because inclusive growth can influence later transformation; omitted country characteristics may affect both sectoral structure and inclusiveness; and past inclusive growth may transmit into current outcomes (Nickell, 1981; Arellano and Bond, 1991; Blundell and Bond, 1998). The BC-MM estimator is therefore preferred because it reduces bias in dynamic panels and is suitable where the dependent variable is persistent. This is particularly important because the inclusive growth index is expected to evolve gradually, reflecting the accumulated effects of welfare, employment and capability conditions rather than only short-run fluctuations.

Finally, the study uses the Common Correlated Effects (CCE) estimator as an additional robustness check because macro-panel data may contain unobserved common factors, even when Cross-Sectional Dependence (CSD) tests are not significant. Consistent with Chudik and Pesaran (2015), CCE absorbs common shocks that can affect countries simultaneously. The median quantile regression with year fixed effects tests whether results are sensitive to outliers and conditional-mean assumptions, while the one-year lagged-regressor model with year fixed effects reduces contemporaneous feedback. Together, the methods provide a layered empirical strategy in which the benchmark estimates establish patterns, BC-MM supplies the main inference and robustness checks assess sensitivity to data, variables and methods.

3.2.1 Bias-corrected method of moments

The primary model is the BC-MM dynamic specification, which regresses inclusive growth on its lag (Breitung et al., 2022), the structural transformation index, sectoral value-added variables, employment shares and controls. The lagged dependent variable captures persistence in inclusive growth, while the structural transformation index captures the overall movement of economic resources across sectors.

Equation (1) shows the dynamic specification given by

(1)

where IGi,t is the index of inclusive growth, α is the intercept, IGi,t1 is the first lag of the dependent variable which allows us to assess the persistence of inclusive growth through dynamic specification and φ is the autoregressive coefficient associated with the lagged dependent variable. STi,t is a measure of structural transformation, while β is its regression coefficient. X is a vector of control variables, and δ is the vector of their associated parameters. ωi is the country-specific effect, and μi,t is a random disturbance representing the idiosyncratic error. The model is estimated in four specifications because multicollinearity is observed among agriculture value added, industry value added, employment in agriculture and services value added. Each specification excludes a different high Variance Inflation Factor (VIF) combination, allowing the analysis to compare results without allowing overlapping sectoral shares to dominate the estimates.

The descriptive statistics in Table 2 show that inclusive growth in SSA is generally weak, with a mean index of −0.33 and moderate dispersion, while the structural transformation index averages about 41 with high variation, reflecting uneven progress across countries. Agriculture and services value-added average about 20% and 46% of GDP, respectively, but their sizeable standard deviations show substantial cross-country and overtime differences in production structures. Employment remains heavily concentrated in agriculture, averaging about 50%, while employment in services and industry is lower but dispersed, indicating gradual yet uneven labour reallocation. Manufacturing and industry value added show moderate averages, pointing to different levels of industrialisation, while human capital, institutions and investment display wide variation. Inflation has a relatively low average but extremely high dispersion, signalling episodes of macroeconomic instability.

Table 2

Descriptive statistics, VIF and Pesaran CSD results

Descriptive statisticsMulticollinearityCross-sectional dependence test
VariableObsMeanStd. devMinMaxVIFPesaran CSD tCSD p-value
Inclusive growth766−0.3290.396−3.0901.200 0.620.536
Structural transformation79841.04913.9191.70072.4002.96−0.170.865
Agriculture VA80019.73911.4261.58055.01411.60−1.160.310
Employment in agriculture80050.31020.3165.31891.9306.161.580.114
Manufacturing VA77210.3614.1451.43124.5581.451.180.236
Employment in services80036.47414.8595.81672.529not reported1.440.149
Employment in industry80013.2217.2522.25539.2524.500.780.435
Services VA79346.3928.76712.49077.0205.97−0.350.728
Industry VA78826.05911.3628.02072.1537.870.100.923
Institutions79843.16412.17018.90075.4002.11−1.040.297
Trade in services74314.6167.5983.14750.1451.40−1.310.191
Inflation76810.60544.502−16.860736.1101.09−0.460.644
Gross fixed capital formation76822.1348.1321.09778.0011.650.330.745
Human capital80023.5468.4661.00049.2002.480.030.976

Note(s): VIF denotes variance inflation factor. The VIF column is not applicable to the dependent variable. Pesaran Cross Sectional Dependence (CSD) tests the null hypothesis of cross-sectional independence

Source(s): Computed by authors using Stata 17

The VIF results in Table 2 show manageable but non-trivial multicollinearity. Most variables have VIF values below conventional thresholds, yet agriculture value added, industry value added, employment in agriculture and services value-added show substantial overlap with related sectoral shares. Accordingly, four specifications are estimated, each omitting a different set of high-VIF variables. The Pesaran cross-sectional dependence results in Table 2 show p-values above the 1%, 5% and 10% significance levels for all variables. Therefore, the null of cross-sectional independence cannot be rejected, implying that there is no statistical evidence that shocks in one SSA country are systematically transmitted to others through the variables in the model.

Correlation results which are presented in Table A1 in the Online Appendix are consistent with the study’s theoretical framing because inclusive growth is positively associated with structural transformation, human capital, institutional quality, trade in services and especially employment in services and industry, while it is negatively associated with agriculture value added and agricultural employment. These patterns suggest that countries moving further toward non-agricultural sectors tend to record more inclusive outcomes, but the correlation matrix also shows strong association among sectoral variables, justifying the multicollinearity-sensitive specifications.

Panel unit-root tests using Fisher-ADF and Fisher-PP procedures presented in the Online Appendix as Table A2 indicate a mixed order of integration. Inclusive growth, structural transformation, institutions, agriculture value added, services value added, industry value added, trade in services, inflation and gross fixed capital formation are stationary in levels, while employment shares and human capital are generally stationary after first differencing, with no evidence of I(2) processes. This mixed I(0) and I(1) pattern supports dynamic panel modelling.

The Pedroni and Kao cointegration tests presented in Table 3 below further indicate that inclusive growth and its structural drivers share a stable long-run path, since most statistics reject the null of no cointegration. Thus, deviations from equilibrium are temporary, justifying long-run interpretation and dynamic estimation. These preliminary results also explain why the study avoids relying on a single static model. The positive correlation between non-agricultural employment and inclusive growth supports the theoretical expectation that reallocation can be beneficial, while the negative correlation with agriculture and agricultural employment signals that persistent dependence on low-productivity activities remains a challenge. However, the VIF results show that these relationships are statistically intertwined, and the unit-root and cointegration results show that short-run variation must be interpreted alongside longer-run adjustment. For that reason, the subsequent regressions are treated as a sequence of increasingly rigorous checks rather than isolated tables (see Table 3).

Table 3

Panel cointegration – Pedroni and Kao results

Statisticp-value
Panel I: Pedroni test for cointegration
Modified Phillips–Perron t2.830**0.0023
Phillips–Perron t−10.315**0.0000
Augmented Dickey–Fuller t−9.635**0.0000
Panel II: Kao test for cointegration
Modified Dickey–Fuller t0.2760.3911
Dickey–Fuller t−1.698**0.0447
Augmented Dickey–Fuller t−2.221**0.0132
Unadjusted modified Dickey–Fuller t−0.7680.2212
Unadjusted Dickey–Fuller t−2.427**0.0076
Source(s): Computed by author using secondary data with Stata output, where ** denotes 5% significance level

The pooled OLS results presented in Table A3 in the Online Appendix show a positive and significant relationship between structural transformation and inclusive growth across all four specifications. Agriculture value added is negative and significant, industry value added is positive and significant in the relevant specification, services value added is negative and significant when entered, and manufacturing value added is consistently insignificant. Employment in services is positive and significant across all specifications, whereas employment in agriculture and industry are negative and significant only in the specification that emphasises employment shares. Trade in services is also positive and significant throughout, while institutions are negative and significant, and investment and human capital are mostly insignificant.

The between-country estimates, Table A4 in the Online Appendix, narrow the robust findings. Employment in services and trade in services remain positive and significant, but structural transformation and most sectoral value-added variables become insignificant, implying that some relationships are stronger within countries over time than between country averages. These benchmark results are useful because they show that services employment and services trade are recurrent correlates of inclusive growth, while sectoral value-added effects vary with model structure.

However, because pooled OLS and between estimates do not fully resolve dynamic bias, omitted heterogeneity or simultaneity, they are treated as supportive rather than decisive. The full benchmark results’ tables are therefore retained in the Online Appendix as Table A3 and Table A4, respectively, and the discussion focuses on their role in motivating the dynamic estimates.

Lagged inclusive growth is positive in all four specifications in column (1), where a one-unit increase in previous inclusive growth raises current inclusive growth by about 0.27 units, indicating persistence. Structural transformation is positive and highly significant in all four specifications. A one-unit increase in the structural transformation index increases inclusive growth by about 0.033 units in column (1), 0.028 units in column (2) and column (3), and roughly 0.033 units in column (4). This stability indicates that, after controlling for dynamic bias and endogeneity, deeper reallocation of economic activity is strongly associated with higher inclusive growth. This result directly supports the first objective and confirms that transformation matters as a broader process through which productivity, demand, employment opportunities and linkages evolve.

The sectoral value-added results are more selective. Agriculture value added appears in column (1) and is positive and highly significant, indicating that a one percentage point increase in agriculture value added raises the inclusive growth index by about 0.043 units. This implies that agriculture can support inclusive growth when embedded in a broader structural transformation process, especially where value addition raises rural productivity and strengthens linkages. By contrast, manufacturing value added is negative but insignificant in all specifications, industry value added is insignificant where included, and services value added is also insignificant where included. The study does not find a robust independent contribution from manufacturing, industry or services value added after dynamics and other controls are accounted for. Instead, their effects appear to operate through the overall structural transformation index, trade in services or employment channels rather than sectoral output shares alone.

With respect to employment shifts, the BC-MM results show no statistically reliable direct contemporaneous effect from employment in agriculture, employment in industry or employment in services. Employment in agriculture in column (3) is insignificant, while employment in industry and services is insignificant across all specifications. A one percentage point change in these employment shares does not significantly change the inclusive growth index once past inclusive growth, structural transformation, sectoral value added and the controls are included. This refines the static results, where service employment appeared important, by showing that in the dynamic setting, the immediate employment-share effects are absorbed by broader transformation and trade channels. The employment effect may depend on job quality, formality, wages and destination-sector productivity, which aggregate employment shares cannot fully measure.

Gross fixed capital formation is negative and highly significant across all BC-MM specifications, with coefficients around −0.040 to −0.044, suggesting that higher aggregate investment is associated with lower inclusive growth over the estimation horizon. Trade in services is strongly positive across all specifications, with a one percentage point increase raising inclusive growth by about 0.022–0.031 units, indicating that services trade is a key channel through which external integration supports inclusion. Inflation is negative only in column (1), while human capital is significant but sign-changing. The AR(2) p-values are well above conventional significance levels, so the null of no second-order serial correlation cannot be rejected.

Table A5, which is presented in the Online Appendix, reports the Dynamic Common Correlated Effects estimates, which explicitly control for unobserved common shocks and cross-sectional dependence, thereby providing a demanding robustness check of earlier results. In column (1), structural transformation remains positive and statistically significant at the 10% level, indicating that a one unit increase in the structural transformation index increases the inclusive growth index by about 0.0185 units. Agriculture value added has a marginally significant negative influence, while employment in services, services trade and inflation are positive and significant. These findings imply that once common shocks are accounted for, structural transformation, services-based labour reallocation and services trade remain visible channels shaping inclusive growth. In column (2), structural transformation remains positive and significant, while the individual sectoral value added and employment variables are insignificant, suggesting that the composite transformation index is more robust than isolated sectoral shares. In columns (3) and (4), most coefficients lose significance, although inflation and human capital are significant in column (3).

Two additional robustness checks were conducted using the same four model specifications with the results presented in Tables A6 and A7 in the Online Appendix. First, median quantile regression with year fixed effects was estimated to test whether the findings are driven by outliers or by the conditional-mean orientation of OLS-type estimators. The results which are moved to the Online Appendix as Table A6 show that structural transformation remains positive and significant in columns (1), (3) and (4), while employment in services and trade in services remain positive and significant when included. Second, a one-year lagged regressor model, with year fixed effects and country-clustered standard errors, was estimated to reduce simultaneity concerns. The results which are moved to the Online Appendix as Table A7 showed that structural transformation remains positive and significant in columns (3) and (4), while lagged employment in services and lagged trade in services retain positive and significant effects. These robustness checks reinforce the main conclusion that structural transformation and service-sector channels remain central to inclusive growth in SSA.

The BC-MM findings reported in Table 4 speak to the central theoretical claim that structural transformation is a fundamental driver of development, but through an explicitly inclusive lens. The positive coefficients on the structural transformation index across all specifications imply that deeper structural reallocation is associated with more inclusive growth after endogeneity and dynamics are accounted for in SSA. This is broadly in line with the classical dual-economy intuition of Lewis (1954) and growth patterns identified by Kuznets (1955), which emphasised movement out of traditional agriculture to modern or formal sector activities. It is consistent with Trofimov (2023), Ogunleye and Favour (2025) and Tiako (2024), who find that structural transformation enhances aggregate growth positively, although short-run effects may be ambiguous.

Table 4

Bias-corrected method of moments results

Variable(1)(2)(3)(4)
Lagged inclusive growth0.2678**0.13710.13000.1341
(0.1286)(0.1670)(0.1638)(0.1540)
Structural transformation0.0327***0.0276***0.0282***0.0332***
(0.0084)(0.0080)(0.0082)(0.0108)
Agriculture VA0.0430***   
(0.0118)   
Industry VA −0.0030  
 (0.0023)  
Manufacturing VA−0.0284−0.0586−0.0592−0.0590
(0.0537)(0.0537)(0.0529)(0.0498)
Services VA   −0.0104
   (0.0068)
Employment in agriculture  −0.0046 
  (0.0559) 
Employment in industry0.15830.03230.0071−0.0765
(0.1591)(0.1697)(0.2237)(0.1933)
Employment in services−0.0454−0.0016omitted0.0317
(0.0478)(0.0562) (0.0638)
Gross fixed capital formation−0.0404***−0.0408***−0.0410***−0.0444***
(0.0051)(0.0033)(0.0036)(0.0070)
Trade in services0.0271***0.0311***0.0297***0.0224***
(0.0017)(0.0018)(0.0011)(0.0021)
Inflation−0.0076**−0.0078−0.0076−0.0064
(0.0033)(0.0053)(0.0056)(0.0051)
Institutions−0.00580.00740.00860.0119
(0.0115)(0.0112)(0.0100)(0.0099)
Human capital0.0152**−0.0087***−0.0090***−0.0069*
(0.0068)(0.0022)(0.0023)(0.0029)
Constant−2.9306***−0.8059−0.3300−0.0502
(0.6925)(0.4933)(6.0007)(0.5848)
Observations595595595590
Instruments13131213
Hansen, p-value0.31450.41930.34770.3926
Countries32323232
AR(1) test, p-value0.0500.0950.1000.107
AR(2) test, p-value0.9880.9810.9550.818

Note(s): ***p < 0.1, **p < 0.05, *p < 0.01. Column (1) excludes industry VA, employment in agriculture and services VA; column (2) excludes agriculture VA, employment in agriculture and services VA; column (3) excludes agriculture VA, industry VA and services VA; Column (4) excludes agriculture VA, industry VA and employment in agriculture. Standard errors are reported in parentheses. Inclusive growth (L1) denotes one-period lagged inclusive growth. AR(1) and AR(2) report the p-values for the null hypothesis of no first and second-order serial correlation, respectively

Source(s): Computed by authors using secondary data with Stata 15 output

The current study contributes that, in addition to raising gross domestic product, structural transformation accelerates growth inclusiveness when evaluated by a multiple-dimensional index. While Trofimov (2023) argues that rural hold-back and excess services expansion are suffocating GDP growth, the positive transformation effect here suggests that for SSA, what changes is important for inclusive growth and whether the transformation index captures movement into more productive and better-connected activities. This is in line with Lin (2011), that development is about creating capabilities based on sectors of latent competitive advantage, and Olaoye et al.(2026), who demonstrate how economic transformation supports wider sustainable development once knowledge and technology are mobilised. Lagged inclusive growth echoes Stawska and Jabłonska (2022), indicating the non-automatic nature of inclusive pathways emphasised within theoretical literature.

The sectoral value added and control variable findings engage with the empirical debate on which sectors and macro conditions drive inclusive development. Agriculture value added is positive and strongly significant in column (1), contradicting classical predictions of agriculture-to-manufacturing transition and findings such as Trofimov (2023), Dabús and Delbianco (2021) and Mateko (2025); however, it is consistent with Enongene (2024), who reports that agricultural, manufacturing, industrial and services value added are poverty reducing. The positive agriculture coefficient, conditional on structural transformation, implies that productivity increases, commercialisation and stronger linkages to non-farm sectors can underpin inclusive growth. This nuance is in line with Hirschman (1958), Kaldor (1966) and Erumban and de Vriesy (2024).

In contrast, the study does not find strong contributions of value added in manufacturing, industry and services, which sits uncomfortably with Kaldor’s manufacturing focus and Asian evidence by Riaz et al. (2020), Lin et al. (2022) and Pham and Riedel (2019). This contrast is explicable with reference to Rodrik’s (2016) thesis of premature deindustrialisation and findings by Magwedere and Marozva (2024) and Mamman and Sohag (2023). Employment shares are also insignificant in the bias-corrected dynamic specifications, aligning with Jobe and Ricciuti (2023), Amadou and Aronda (2020) and Aggarwal (2018a, b). As such, the most robust short-run drivers of inclusive growth are the quality of structural transformation, the nature of sectoral value added and integration into services trade, while employment shifts, human capital and institutions matter through slower moving channels.

This study examined how sectoral value added and employment shifts drive inclusive growth in SSA using annual unbalanced panel data for 32 countries over 2000–2024. The central conclusion is that structural transformation has a positive and statistically significant effect on inclusive growth. In the main BC-MM estimates, the structural transformation coefficient remains positive across all four specifications, ranging from about 0.028 to 0.033, indicating that broad-based reallocation of economic activity improves inclusive growth when dynamic bias and endogeneity are addressed. This finding supports policies that treat structural transformation as a core inclusive-growth strategy rather than as a narrow sectoral accounting exercise.

However, the sectoral findings show that not all forms of transformation are equally inclusive. Agriculture value added is positive and significant, with a coefficient of about 0.043, whereas manufacturing, industry and services value added do not show robust independent effects in the dynamic model. This implies that productivity-enhancing agriculture remains central to inclusion in SSA, especially where it raises rural incomes and strengthens linkages with agro-processing, trade and services. Therefore, governments should not abandon agriculture in pursuit of industrialisation but should modernise it through productivity, value-chain development and market access. At the same time, manufacturing and industry policies should focus on employment intensity and domestic linkages, because higher output shares alone are insufficient.

The employment-shift results further show that aggregate sectoral employment shares have insignificant direct effects once structural transformation and other covariates are controlled for. Policy should therefore prioritise the quality of employment rather than only the movement of labour across sectors. For services, the evidence suggests that employment absorption is potentially important, but it must be accompanied by productivity, formalisation and skills upgrading. Finally, since trade in services is consistently positive, SSA countries should strengthen tradable services, digital services, transport, tourism and business services while aligning these with human capital and infrastructure policy. Detailed benchmark and robustness results are retained in the Online Appendix, while the main article emphasises the core finding. That is, inclusive structural transformation requires productivity growth, employment quality and sectoral linkages to move together.

In the process, one potential limitation is that of availability and quality of sectoral employment data and data on value added across SSA countries at high frequency with consistent classification in terms of the sectors which might affect how granularly the analysis can be carried out. Furthermore, although the composite indices of inclusive growth and structural transformation were developed using state-of-the-art techniques, they might not record all the dimensions and complexities of these multifaceted phenomena.

Next steps could involve studying alternative methods of analysis (e.g. panel threshold models or quantile regressions) in order to draw further conclusions with respect to the non-linear relationship between the process of structural transformation and inclusive growth. Second, taking a finer disaggregated perspective of the services sector in terms of high-productivity modern and low-productivity traditional services could offer a richer insight into inclusivity determinants. Future research might also replicate this dual-channel conceptualisation over more developing countries or on specific sub-regions in Africa to ascertain the generalisability of the results and compare contextual differences in the structural transformation-inclusive growth (ST-IG) relationship.

The supplementary material for this article can be found online.

AfDB
(
2015
),
Poverty, Inequality and Employment in Africa
,
African Development Bank
,
Abidjan
.
Aggarwal
,
N.
(
2018a
), “
Growth, structural change and employment linkages in India
”,
World Development
, Vol. 
112
, pp. 
1
-
15
.
Aggarwal
,
A.
(
2018b
), “
Economic growth, structural change and productive employment linkages in India: did market transition matter?
”,
South Asia Economic Journal
, Vol. 
19
No. 
1
, pp. 
64
-
85
, doi: .
Amadou
,
A.
and
Aronda
,
T.
(
2020
), “
Structural transformation in Sub-Saharan Africa: a comparative analysis of sub-regions performances
”,
African Journal of Economic and Management Studies
, Vol. 
11
No. 
2
, pp. 
233
-
252
, doi: .
Arellano
,
M.
and
Bond
,
S.
(
1991
), “
Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations
”,
The Review of Economic Studies
, Vol. 
58
No. 
2
, pp. 
277
-
297
, doi: .
Asongu
,
S.
and
Odhiambo
,
N.
(
2019
), “
Financial access, governance and insurance sector development in Sub-Saharan Africa
”,
Journal of Economic Studies
, Vol. 
47
No. 
4
, pp. 
849
-
875
, doi: .
Asongu
,
S.
,
Biekpe
,
N.
and
Tchamyou
,
V.
(
2018
), “
Remittances, ICT and doing business in Sub-Saharan Africa
”,
Journal of Economic Studies
, Vol. 
46
No. 
1
, pp. 
35
-
54
, doi: .
Atangana
,
S.S.
,
Nzepang
,
F.
and
Nguenda Anya
,
S.B.
(
2025
), “
What interaction between income inequality and structural transformation in Sub-Saharan Africa?
”,
The Journal of Economic Inequality
, Vol. 
23
No. 
4
, pp. 
1387
-
1410
, doi: .
Baltagi
,
B.H.
(
2008
),
Econometric Analysis of Panel Data
, (4th ed.) ,
John Wiley & Sons
,
Chichester
.
Basu
,
M.
and
Nag
,
R.N.
(
2020
), “
Open economy macroeconomics of commodity price fluctuation, sectoral inter-linkage and employment
”,
Journal of Economic Studies
, Vol. 
47
No. 
6
, pp. 
1467
-
1494
, doi: .
Baymul
,
C.
and
Sen
,
K.
(
2019
), “
Was kuznets right? New evidence on the relationship between structural transformation and inequality
”,
The Journal of Development Studies
, Vol. 
56
No. 
9
, pp. 
1643
-
1662
, doi: .
Blundell
,
R.
and
Bond
,
S.
(
1998
), “
Initial conditions and moment restrictions in dynamic panel data models
”,
Journal of Econometrics
, Vol. 
87
No. 
1
, pp. 
115
-
143
, doi: .
Breitung
,
J.
,
Kripfganz
,
S.
and
Hayakawa
,
K.
(
2022
), “
Bias-corrected method of moments estimators for dynamic panel data models
”,
Econometrics and Statistics
, Vol. 
24
, pp. 
140
-
156
,
[e-journal]
doi: .
Buera
,
F.J.
and
Kaboski
,
J.P.
(
2012
), “
The rise of the service economy
”,
American Economic Review
, Vol. 
102
No. 
6
, pp. 
2540
-
2569
, doi: .
Busse
,
M.
,
Erdogan
,
C.
and
Mühlen
,
H.
(
2019
), “
Structural transformation and its relevance for economic growth in Sub-Saharan Africa
”,
Review of Development Economics
, Vol. 
23
No. 
4
, pp. 
1693
-
1718
, doi: .
Cadot
,
O.
,
Rollo
,
V.
and
Sáez
,
S.
(
2016
),
Why is Structural Transformation so Slow in Sub-saharan Africa?
,
The World Bank
,
Washington, DC
.
Chenery
,
H.B.
and
Taylor
,
L.
(
1968
), “
Development patterns: among countries and over time
”,
The Review of Economics and Statistics
, Vol. 
50
No. 
4
, p.
391
, doi: .
Choi
,
I.
(
2001
), “
Unit root tests for panel data
”,
Journal of International Money and Finance
, Vol. 
20
No. 
2
, pp. 
249
-
272
, doi: .
Chudik
,
A.
and
Pesaran
,
M.H.
(
2015
), “
Common correlated effects estimation of heterogeneous dynamic panel data models with weakly exogenous regressors
”,
Journal of Econometrics
, Vol. 
188
No. 
2
, pp. 
393
-
420
, doi: .
Comin
,
D.
,
Lashkari
,
D.
and
Mestieri
,
M.
(
2021
), “
Structural change with long-run income and price effects
”,
Econometrica
, Vol. 
89
No. 
1
, pp. 
311
-
374
, doi: .
Dabús
,
C.
and
Delbianco
,
F.
(
2021
), “
Economic development and structural change: the role of services
”,
Journal of Economic Development
, Vol. 
46
No. 
3
, pp. 
65
-
84
.
Diao
,
X.
,
McMillan
,
M.
and
Rodrik
,
D.
(
2019
), “The recent growth boom in developing economies: a structural-change perspective”, in
The Palgrave Handbook of Development Economics
, pp. 
281
-
334
, doi: .
Driscoll
,
J.C.
and
Kraay
,
A.C.
(
1998
), “
Consistent covariance matrix estimation with spatially dependent panel data
”,
Review of Economics and Statistics
, Vol. 
80
No. 
4
, pp. 
549
-
560
, doi: .
Enongene
,
B.E.
(
2024
), “
Structural transformation and poverty alleviation in Sub-Saharan Africa countries: sectoral value-added analysis
”,
Journal of Business and Socio-Economic Development
, Vol. 
4
No. 
4
, pp. 
326
-
339
, doi: .
Erumban
,
A.A.
and
de Vries
,
G.J.
(
2024
), “
Structural change and poverty reduction in developing economies
”,
World Development
, Vol. 
181
, 106674, doi: .
Gamette
,
P.
,
Odhiambo
,
N.M.
and
Asongu
,
S.
(
2025
), “
The role of governance in remittances-access to electricity nexus in Sub-Saharan Africa? A rural-urban comparative analysis
”,
Journal of Economic Studies
, Vol. 
52
No. 
9
, pp. 
69
-
88
, doi: .
Gollin
,
D.
,
Jedwab
,
R.
and
Vollrath
,
D.
(
2016
), “
Urbanization with and without industrialization
”,
Journal of Economic Growth
, Vol. 
21
No. 
1
, pp. 
35
-
70
, doi: .
Herrendorf
,
B.
,
Rogerson
,
R.
and
Valentinyi
,
Á.
(
2013
), “Growth and structural transformation”, in
Aghion
,
P.
 and
Durlauf
,
S.N.
(Eds),
Handbook of Economic Growth
,
Elsevier
,
Amsterdam
, Vol. 
2
, pp. 
855
-
941
.
Hirschman
,
A.O.
(
1958
),
The Strategy of Economic Development
,
Yale University Press
,
New Haven
.
Jobe
,
A.M.
and
Ricciuti
,
R.
(
2023
), “
Sectoral shifts and labour market outcomes in Sub-Saharan Africa
”,
UNU-WIDER Working Paper 138/2023, United Nations University World Institute for Development Economics Research (UNU-WIDER), Helsinki
, doi: .
Kagochi
,
J.
(
2019
), “
Inflation and financial sector development in sub-Saharan African countries
”,
Journal of Economic Studies
, Vol. 
46
No. 
4
, pp. 
798
-
811
, doi: .
Kaldor
,
N.
(
1966
),
Causes of the Slow Rate of Economic Growth of the United Kingdom
,
Cambridge University Press
,
Cambridge
.
Kongsamut
,
P.
,
Rebelo
,
S.
and
Xie
,
D.
(
2001
), “
Beyond balanced growth
”,
The Review of Economic Studies
, Vol. 
68
No. 
4
, pp. 
869
-
882
, doi: .
Kuznets
,
S.
(
1955
), “
Economic growth and income inequality
”,
The American Economic Review
, Vol. 
45
No. 
1
, pp. 
1
-
28
.
Lewis
,
W.A.
(
1954
), “
Economic development with unlimited supplies of labour
”,
The Manchester School
, Vol. 
22
No. 
2
, pp. 
139
-
191
, doi: .
Lin
,
J.Y.
(
2011
),
New Structural Economics: a Framework for Rethinking Development and Policy
,
World Bank
,
Washington, DC
.
Lin
,
J.Y.
,
Wang
,
Y.
and
Xu
,
J.
(
2022
), “
Rethinking industrial policy from the perspective of new structural economics
”,
Development Policy Review
, Vol. 
40
No. 
5
, e12609, doi: .
Loayza
,
N.V.
and
Raddatz
,
C.
(
2010
), “
The composition of growth matters for poverty alleviation
”,
Journal of Development Economics
, Vol. 
93
No. 
1
, pp. 
137
-
151
, doi: .
Maddala
,
G.S.
and
Wu
,
S.
(
1999
), “
A comparative study of unit root tests with panel data and a new simple test
”,
Oxford Bulletin of Economics and Statistics
, Vol. 
61
No. 
Supplement 1
, pp. 
631
-
652
, doi: .
Magwedere
,
M.R.
and
Marozva
,
G.
(
2024
), “
Structural transformation and inequality: a sectoral analysis for low- and middle-income countries
”,
Review of Economic Analysis
, Vol. 
16
No. 
4
, pp. 
423
-
441
, doi: .
Mamman
,
S.O.
and
Sohag
,
K.
(
2023
), “
Inclusive growth and structural transformation: the role of innovation and digitalisation spillover
”,
Ekonomika regiona/Economy of Regions
, Vol. 
19
No. 
3
, pp. 
598
-
611
, doi: .
Marfatia
,
H.
(
2023
), “
The financial market's ability to forecast economic growth: information from sectoral movements
”,
Journal of Economic Studies
, Vol. 
50
No. 
7
, pp. 
1467
-
1484
, doi: .
Mateko
,
F.M.
(
2025
), “
The nexus between structural transformation and poverty reduction in Maghreb region: a sectoral value-added analysis
”,
Revista de Economia Política/Brazilian Journal of Political Economy
, Vol. 
45
No. 
2
, e253646, doi: .
McMillan
,
M.S.
and
Rodrik
,
D.
, (
2011
), “
Globalization, structural change and productivity growth
”,
NBER Working Paper No. 17143. Cambridge, MA: National Bureau of Economic Research
.
McMillan
,
M.S.
,
Rodrik
,
D.
and
Verduzco-Gallo
,
Í.
(
2014
), “
Globalization, structural change, and productivity growth, with an update on Africa
”,
World Development
, Vol. 
63
, pp. 
11
-
32
, doi: .
Mlachila
,
M.
,
Tapsoba
,
R.
and
Tapsoba
,
S.J.A.
(
2016
), “
A quality of growth index for developing countries: a proposal
”,
Social Indicators Research
, Vol. 
134
No. 
2
, pp. 
675
-
710
, doi: .
Morsy
,
H.
,
Shimeles
,
A.
and
Nabassaga
,
T.
(
2023
), “
Structural change and inequality in Africa
”,
Journal of African Economies
, Vol. 
32
No. 
Supplement_2
, pp. 
ii228
-
ii245
, doi: .
Ngai
,
L.R.
and
Pissarides
,
C.A.
(
2007
), “
Structural change in a multisector model of growth
”,
American Economic Review
, Vol. 
97
No. 
1
, pp. 
429
-
443
, doi: .
Nickell
,
S.
(
1981
), “
Biases in dynamic models with fixed effects
”,
Econometrica
, Vol. 
49
No. 
6
, pp. 
1417
-
1426
, doi: .
Ogunleye
,
E.O.
and
Favour
,
O.T.
(
2025
), “
The effects of structural change on economic growth in ECOWAS countries
”,
International Journal of Multidisciplinary Research and Growth Evaluation
, Vol. 
6
No. 
1
, pp. 
673
-
678
.
Olaoye
,
O.O.
,
Zerihun
,
M.F.
and
Tabash
,
M.I.
(
2026
), “
Structural transformation and sustainable development in Sub-Saharan Africa
”,
African Journal of Economic and Management Studies
, Vol. 
17
No. 
3
, pp. 
383
-
404
, doi: .
Pham
,
T.H.
and
Riedel
,
J.
(
2019
), “
Impacts of the sectoral composition of growth on poverty reduction in Vietnam
”,
Journal of Economics and Development
, Vol. 
21
No. 
2
, pp. 
213
-
222
, doi: .
Rashid
,
A.
and
Intartaglia
,
M.
(
2017
), “
Financial development, financial inclusion and poverty reduction: evidence from developing countries
”,
Journal of Economic Studies
, Vol. 
44
No. 
4
, pp. 
654
-
669
, doi: .
Riaz
,
U.
,
Hussain
,
B.
and
Bhatti
,
A.A.
(
2020
), “
Role of structural change in economic growth: an empirical evidence from a panel of Asian countries
”,
Liberal Arts and Social Sciences International Journal (LASSIJ)
, Vol. 
4
No. 
2
, pp. 
345
-
360
, doi: .
Rodrik
,
D.
(
2016
), “
Premature deindustrialization
”,
Journal of Economic Growth
, Vol. 
21
No. 
1
, pp. 
1
-
33
, doi: .
Ssozi
,
J.
,
Asongu
,
S.
and
Amavilah
,
V.H.
(
2019
), “
The effectiveness of development aid for agriculture in Sub-Saharan Africa
”,
Journal of Economic Studies
, Vol. 
46
No. 
2
, pp. 
284
-
305
, doi: .
Stawska
,
J.
and
Jabłońska
,
M.
(
2022
), “
Determinants of inclusive growth in the context of the theory of sustainable finance in the European union countries
”,
Sustainability
, Vol. 
14
No. 
1
, 100, doi: .
Tiako
,
I.
(
2024
), “
Is structural transformation generate economic growth in Sub-Saharan Africa
”,
Theoretical Economics Letters
, Vol. 
14
No. 
3
, pp.
761
-
784
, doi: .
Timmer
,
M.P.
and
De Vries
,
G.J.
(
2008
), “
Structural change and growth accelerations in Asia and Latin America: a new sectoral data set
”,
Cliometrica
, Vol. 
3
No. 
2
, pp. 
165
-
190
, doi: .
Trofimov
,
I.D.
(
2023
), “
The effects of structural change on economic growth: a panel data analysis
”,
Panoeconomicus
, Vol. 
73
No. 
2
, pp. 
129
-
162
, doi: .
United Nations Development Programme (UNDP)
(
2022
),
Human Development Report 2021/2022: Uncertain Times, Unsettled Lives – Shaping our Future in a Transforming World
,
UNDP
,
New York
.
Wooldridge
,
J.M.
(
2010
),
Econometric Analysis of Cross Section and Panel Data
, (2nd ed.) ,
MIT Press
,
Cambridge, MA
.
World Bank
(
2022
),
World Development Report 2022: Finance for an Equitable Recovery
,
The World Bank
,
Washington, DC
.
Yerrabati
,
S.
(
2024
), “
Path to achieving SDG 8: do worker remittances reduce vulnerable employment?
”,
Journal of Economic Studies
,
[Preprint]
, doi: .
Anand
,
R.
,
Mishra
,
S.
and
Peiris
,
S.J.
(
2013
), “
Inclusive growth: measurement and determinants
”,
IMF Working Paper No. 13/135, International Monetary Fund, Washington, DC
, Vol. 
2013
No. 
135
, p.
1
, doi: .
World Bank
(
2024
),
World Development Indicators
,
World Bank
,
Washington, DC
,
available at:
 Link to the website
Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at Link to the terms of the CC BY 4.0 licence.

Supplementary data

or Create an Account

Close Modal
Close Modal