Skip Content

How civil service reform can curb corruption in developing countries

Blog • • Jan-Hinrik Meyer-Sahling, Christian Schuster, Kim Sass Mikkelsen

Civil service reform is a key instrument to tackle corruption in developing countries. The British Academy/DfID-funded project ‘Civil Service Reform and Anti-Corruption in Developing Countries’ aimed to identify how civil service management practices affect corruption, clientelism and performance of civil servants in developing countries. Civil service reforms to-date have a poor track record, and a limited existing evidence base on ‘what works in civil service management in developing countries’. To help address this evidence gap, the project surveyed 23,000 civil servants in ten countries in four developing regions – the largest original comparative cross-country survey of civil servants ever conducted. The resulting data provides a much-improved evidence basis for the design of civil service reform programmes in developing countries. 

March against corruption - Florianópolis - Brazil

March against corruption - Florianópolis - Brazil. Image credit: Dircinha Welter / Contributor / Getty Images.

The surveyed countries include Ghana, Uganda and Malawi in Africa, Bangladesh and Nepal in Asia, Brazil and Chile in Latin America, and Estonia, Kosovo and Albania in Eastern Europe.

Jan-Hinrik Meyer-Sahling and Christian Schuster present their research at the World Bank

Jan-Hinrik Meyer-Sahling and Christian Schuster present their research at the World Bank.

The team was keen to ensure our learning reached those in decision-making powers. Our comparative, cross-country report has since been presented to international organisations such as the World Bank, the Inter-American Development Bank and the European Commission. The project team has also produced country reports that have been presented to national governments in several countries surveyed.

Connections matter

The survey of civil servants has provided a wealth of insights into civil service management practices, both in individual countries and across countries.

On average across countries, 25% of civil servants at the managerial level indicated that knowing a politician or a person with a political link has helped them to obtain their job. This compared to 22% of administrative-level civil servants, and 18% for technical professional level civil servants. However, the share of civil servants for whom political connections matter varies significantly between countries, from 5%-6% in Estonia to 39%-44% in Kosovo. 

On average across countries, 25% of civil servants at the managerial level indicated that knowing a politician or a person with a political link has helped them to obtain their job.

On average across the countries surveyed, 25% of civil servants at the managerial level indicated that knowing a politician or a person with a political link has helped them to obtain their job. Image credit: Marc Gutierrez / Getty Images.

On average personal connections matter across almost all surveyed institutions and at all levels of hierarchy to get jobs, promotions and pay rises. 41% of civil servants got their first job (at least in part) thanks to support from friends, family and other personal connections; for 34% of civil servants, such personal connections were at least somewhat important for obtaining promotions; and for 22% they mattered for pay rises.  Again, we observed variations in the importance of personal connections across countries. 76% of civil servants surveyed in Nepal said personal connections were important in recruiting civil servants, but this figure was just 19% among mainly managerial civil servants in Brazil.

Getting the basics right

Statistical analyses of the survey data allow us to identify the impact of civil service management practices on performance, work motivation, clientelism and several forms of corruption.

There are three major lessons for international and national level civil service reformers, which could create more motivated, committed and/or less corrupt civil services across the world.

  1. Consistently recruiting on the grounds of merit through set procedures would curb politicisation and nepotism. Jobs should be publicly advertised and candidates should be systematically assessed in written and/or oral examinations. Both civil service politicisation and nepotism are consistently associated with poor performance and corruption outcomes. Merit recruitment procedures help to reduce this and nepotism, thereby enhancing the performance and integrity among civil servants in developing countries.
  2. Performance-oriented promotion, salary and job stability decisions are associated with greater performance and motivation of civil servants in developing countries. Performance evaluation systems can foster this, but only if implemented properly, that is, with objectives set in advance and results tied in some way to promotion, pay or dismissal prospects. If implemented poorly, performance evaluations have the opposite effect and demotivate staff. Reformers should only introduce performance management systems where they believe they can implement them well.
  3. Paying public servants enough would avoid large-scale turnover of motivated personnel. We did not find that higher pay directly leads to greater motivation and less corruption among existing staff. However, the data shows that better performing staff have access to better-paid job opportunities in the private sector and a greater intention to leave the public sector. Selectively targeting pay increases to high performers to retain them can thus play an important role in safeguarding civil service performance.  

Our data points to the importance of getting three ‘basics’ right. At the same time, we would caution strongly against indiscriminate best practice ‘template’ approaches to civil service reform. Our survey suggests that institutions within countries vary sharply in human resource management practices, and in the attitudes and behaviour of their staff. Some institutions may have performance evaluation systems which work well but have poor recruitment systems, while the opposite is the case for others, even within the same country. Some institutions can have motivated staff but have a retention problem, while others in the same country principally struggle with unethical behaviour among staff.

Civil service reforms therefore need to take into account the attitudes and behaviours of staff and of management practices in a given state institution. It is true that central civil service agencies can play important roles in coordinating the collection of evidence and in advancing the three ‘basics’ across state institutions. Yet our analysis demonstrates that civil service reforms need to be tailored to individual state institutions rather than, as in the conventional civil service reform approach, to the public sector as a whole.


Jan-Hinrik Meyer-Sahling is Professor of Political Science at the University of Nottingham and Principal Investigator on the British Academy/DFID-funded Anti-Corruption Evidence project 'Civil Service Reform and Anti-Corruption in Developing Countries'. Christian Schuster is Lecturer in Public Management at University College London and is the project’s Co-Investigator. Kim Sass Mikkelsen is Assistant Professor at the University of Southern Denmark and is Research Associate on the project.

The British Academy-DfID Anti-Corruption Evidence Programme is a £3.6 million initiative which was launched by the British Academy in partnership with the UK Department for International Development in 2015. So far, the Programme has supported eight international research teams working to identify the most successful ways of addressing corruption in developing countries. 

Blog • • Jan-Hinrik Meyer-Sahling, Christian Schuster, Kim Sass Mikkelsen