To Elect or Not to Elect: Leaders, Alternation in Power and Social Welfare in Sub-Saharan Africa

Abstract

This paper investigates whether and how multiparty elections, introduced in many African countries since the early 1990s, affect a government’s commitment to welfare policies. We hypothesise that contested multiparty elections and turnovers between different leaders and political forces in government – even when democratic standards are not met – positively impact the promotion of social welfare. We test these hypotheses through a cross-sectional and time-series research design, making use of our new, comprehensive ‘Africa Leadership Change’ (ALC) dataset. Empirical results confirm that leaders elected through multiparty elections and countries that experience political alternations in government are associated with higher levels of social welfare.

Acknowledgements

This work is part of a research project on ‘The Economic, Social and Political Consequences of Democratic Reforms: A Quantitative and Qualitative Comparative Analysis’ (COD), funded by a Starting Grant of the European Research Council (‘Ideas’, 7^th Framework Programme of the EU). Previous versions of this paper were presented at the Annual Conference of the European Political Science Association (EPSA), Edinburgh, 19–21 June 2014; at a workshop on ‘Consolidating democratic transitions’, Department of International Development, London School of Economics, London, 14–15 July 2014; at the Annual Conference of South African Association of Political Science (SAAPS), University of South Africa, Pretoria, 11–13 September 2014; and at the XXVIII Annual Conference of Italian Political Science Association (SISP), Università degli Studi di Perugia, 11–13 September 2014. We are grateful to E. A. Brett, Andrea Cassani, Fabio Franchino and two anonymous reviewers for useful comments on previous versions of the paper. Data and codes to replicate the analyses conducted in this study are available from the authors upon request.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The redistributive implications of democratic politics are formally theorised in the classic work by Meltzer and Richard (1981).

2. All data in this section are from the ALC dataset, unless otherwise specified.

3. For a more detailed description of the structure, contents and the main sources of ALC, as well as its advantages in researching leadership changes dynamics in Africa compared to other datasets see Carbone and Pellegata (in press).

4. Political regimes are considered democratic if the Polity score is equal to or higher than six (compare with Marshall & Cole, 2011). While some works require a score of seven or more on the −10 to + 10 Polity scale to classify countries as democratic, we believe that a score of six signals a remarkable democratic achievement in an African context that has been historically dominated by authoritarian rule.

5. Models that have HEALTH_SPENDING as dependent variable do not include POST_1990 as a control because health expenditure data are available only from 1995 on. We also excluded NET_ODA_CAPITA from the models presented in Table 2 because HEALTH_SPENDING already includes grants and donations from international agencies and nongovernmental organizations. This allow us to avoid a potential problem of endogeneity. When we include NET_ODA_CAPITA in our models, empirical results do not significantly change.

6. Results of models including variables lagged in other time periods do not substantially change the main findings of our analyses.

Additional information

Funding

This work was supported by the European Research Council [Grant number 262873].