Navigating poverty in developing nations: unraveling the impact of political dynamics on sustainable well-being

The study addresses the pressing challenge of poverty in developing nations, where large shares of populations face income and health deprivations. Building on global agendas (MDGs and SDGs) and economic theories linking institutions to welfare, the authors investigate how political dynamics—corruption, democracy, governance, and political globalization—shape poverty outcomes. The research posits that democracies may enhance public goods and redistribution, while corruption diverts resources and exacerbates inequality; political globalization can either alleviate or worsen poverty depending on channels. The core hypothesis (H1) asserts that political factors have significant impacts on both income and human health poverty in developing countries. To capture multidimensional poverty, the study constructs dual indices and applies advanced panel econometrics to identify dynamic and robust relationships over 1997–2022.

The review synthesizes evidence on political determinants of poverty. Studies report mixed democracy–poverty links (e.g., weak or no direct effects in parts of Sub-Saharan Africa), complex roles of globalization (some reductions in poverty yet potential for rising inequality and health risks), and consistent findings that corruption harms the poor via regressive burdens and resource diversion. Governance quality often correlates with lower poverty, though effects can be context-dependent, non-linear, and stronger in middle-income settings. Prior work frequently uses limited timeframes, narrower geographic coverage, and often does not jointly consider income and health poverty dimensions, motivating the present, more comprehensive approach using recent data, dual poverty indices, and multiple estimators.

Theoretical grounding draws on inclusive institutions (Acemoglu and Robinson), redistribution theory, and the corruption–inequality–poverty linkage (Gupta et al.). Data cover 24 developing countries, annually from 1997–2022. Dependent variables are composite indices built via PCA: Income Poverty Index (IPI) from poverty headcount ratio at $2.15/day (2017 PPP) and poverty gap; Human Health Poverty Index (HHPI) from infant and under-5 child mortality rates. Key independent variables: corruption (Corruption Perceptions Index), democracy (Freedom House Political Rights), governance (institutional quality index combining government effectiveness, control of corruption, voice and accountability, political stability/absence of violence, regulatory quality, rule of law), and political globalization (KOF index). Sources include World Development Indicators, Worldwide Governance Indicators, Freedom House, Transparency International, and KOF. Models: dynamic panels with lagged dependent variables. Baseline specifications: IPI_it = α + α1 IPI_it-1 + γ1 ln(CORRUP)_it + γ2 ln(DEMOC)_it + γ3 GOV_it + γ4 ln(PGLOB)_it + ε_it; HHPI_it analogously. Estimation: Fixed Effects, System GMM (Arellano–Bond/Blundell–Bond; implemented via xtabond2), and 2SLS, with Driscoll–Kraay standard errors to address cross-sectional dependence (CD). Diagnostics include Pesaran CD tests, slope heterogeneity (Pesaran–Yamagata), second-generation unit root tests (CIPS), Pedroni panel cointegration tests, and GMM validity checks (AR(2), Sargan overidentification). Software: EViews 13 and Stata 17. PCA diagnostics: IPI first component eigenvalue 1.9028 (KMO 0.620); HHPI eigenvalue 1.9970 (KMO 0.670); governance index first component eigenvalue 3.7326 (KMO 0.762). Sample size for regressions: 624 observations (24 countries × 26 years).

• Persistence: Lagged poverty is positive and significant, indicating strong persistence. For IPI, SGMM lag coefficient ≈ 0.552 (p<0.01) and 2SLS ≈ 1.125 (p<0.10). For HHPI, SGMM ≈ 1.059 and 2SLS ≈ 1.013 (both p<0.01).
• Corruption: Positive and significant across methods for both poverty dimensions. IPI: FE/DK ≈ 0.291 (p≤0.05), SGMM ≈ 1.034 (p<0.05), 2SLS ≈ 1.508 (p<0.01). HHPI: FE/DK ≈ 1.200 (p<0.01), SGMM ≈ 0.158 (p<0.01), 2SLS ≈ 1.416 (p<0.01).
• Democracy: Significantly reduces income poverty; mixed/mostly insignificant for human health poverty. IPI: FE/DK ≈ −0.127 (p<0.05), SGMM ≈ −1.296 (p<0.01), 2SLS ≈ −0.164 (p<0.01). HHPI: FE ≈ −0.355 (ns), DK ≈ −0.355 (p<0.01), SGMM ≈ −0.030 (ns), 2SLS ≈ −0.292 (p<0.10).
• Governance (institutional quality): Generally positive association with poverty indices, with significance varying by estimator. IPI: FE ≈ 0.215 (p<0.05), DK ≈ 0.215 (p<0.01), SGMM ≈ 0.682 (p<0.10), 2SLS ≈ 0.234 (p<0.05). HHPI: FE ≈ 0.052 (ns), DK ≈ 0.052 (p<0.01), SGMM ≈ 0.031 (p<0.05), 2SLS ≈ 0.070 (ns).
• Political globalization: Negative and significant for both poverty dimensions in most estimators. IPI: FE/DK ≈ −3.779 (p<0.01), SGMM ≈ −3.116 (p<0.01), 2SLS ≈ −3.697 (p<0.05). HHPI: FE/DK ≈ −1.821 (p<0.01), SGMM ≈ −1.220 (ns), 2SLS ≈ −1.715 (p<0.05).
• Panel properties and validity: Evidence of cross-sectional dependence and slope heterogeneity; variables stationary at level or first difference (CIPS); Pedroni tests indicate cointegration. GMM diagnostics show no second-order autocorrelation (AR(2) p>0.05) and valid instruments (Sargan p>0.10).

The results substantiate the hypothesis that political factors materially shape poverty outcomes in developing countries. Corruption consistently exacerbates both income and health-related poverty, aligning with theories that corrupt practices divert public resources, depress service delivery, and erode social trust. Democracy is associated with lower income poverty, consistent with redistribution and public goods provision mechanisms, though its effect on health poverty is weaker or estimator-dependent, indicating that democratic institutions may alleviate poverty primarily through income and service channels rather than immediate health outcomes. The governance index shows a positive association with poverty, with significance varying by method—suggesting that, in this sample and period, prevailing institutional quality may be insufficient or that governance–poverty dynamics are non-linear and context-specific, potentially requiring thresholds before beneficial effects emerge. Political globalization generally reduces both poverty dimensions, consistent with the benefits of international policy linkages, information exchange, and cooperative frameworks. Overall, the dynamic tests confirm persistence in poverty, reinforcing concerns about poverty traps and the need for sustained policy interventions. These findings inform policy by emphasizing anti-corruption measures, strengthening democratic processes, and leveraging international cooperation to reduce poverty, while recognizing that governance reforms may need to reach certain quality thresholds and be tailored to country context to produce consistent poverty reductions.

The study develops dual poverty indices (income and human health) and a governance index via PCA and applies advanced panel estimators (FE, Driscoll–Kraay, System GMM, 2SLS) to 24 developing countries from 1997–2022. It demonstrates that corruption increases, and political globalization decreases, both income and health poverty; democracy significantly reduces income poverty but shows mixed effects on health poverty; governance exhibits a positive association with poverty with estimator-dependent significance, highlighting complex institutional dynamics. Contributions include: (1) quantifying the dynamic impacts of multiple political factors on multidimensional poverty using recent, long-span data; (2) addressing cross-sectional dependence, endogeneity, and heterogeneity in a unified empirical framework; and (3) providing robust, policy-relevant evidence. Future research could examine non-linear and threshold effects of governance and democracy, explore regional heterogeneity, incorporate broader poverty measures (e.g., multidimensional poverty indices), and investigate mechanisms (public spending composition, service delivery, health systems) that mediate political–poverty linkages.

Limitations are not extensively discussed but include: reliance on proxies and composite indices (CPI, Freedom House PR, KOF, PCA-constructed indices) that may introduce measurement error; a sample restricted to 24 developing countries selected by data availability, which may limit generalizability; mixed significance for governance and democracy in the human health model suggests potential non-linearities or threshold effects not fully modeled; despite using SGMM and Driscoll–Kraay to address endogeneity and cross-sectional dependence, residual unobserved heterogeneity and instrumentation limitations may persist.