About me

Contact me

Bureaucracy Blog

My name is Dan Rogger and this is my blog on public administration (you can find out more about me at my About me page of my main web site). It aggregates the various blogs and articles I have done elsewhere on public administration, rather than presenting anything new.  Please send your thoughts and replies to drogger@worldbank.org.


Postings (in reverse chronological order)

Autonomy, incentives, and the effectiveness of bureaucrats | VoxDev
4th September 2017
Co-authored with Imran Rasul and Martin Williams

The capabilities of the state shape the process of economic development. In studying the state, economists have long paid attention to the selection and incentives of both top-tier politicians and frontline workers, such as teachers, tax inspectors, and health workers. What has been relatively less studied is the vital middle-tier of civil servants, typically based in government ministries, whose day-to-day job is converting political decisions into actual policy (Finan et al. 2016). This is surprising given that this middle tier of civil servants is a key human resource and ultimately plays an important role in determining bureaucratic effectiveness and state capabilities.

Analysing management practices and performance

Our recent work tries to make progress on understanding what drives the effectiveness of bureaucrats in two important contexts in sub-Saharan Africa: Nigeria and Ghana (Rasul and Rogger 2016, Rasul et al. 2017). Working with the central civil services in each country, we have been examining whether the management practices that bureaucrats operate under correlate to the effective delivery of public services that they are responsible for.

In the Nigerian context, we hand-coded independent engineering assessments of project completion rates for 4,700 public projects (Rasul and Rogger 2016). For each of 63 civil service organisations tasked to implement these projects (including 10 ministries and 53 other Federal civil service organisations), we surveyed senior bureaucrats to elicit management practices in place for middle-tier bureaucrats, building on methods pioneered by Bloom and Van Reenen (2007) and Bloom et al. (2012).

Our more recent paper is a scientific replication of this work in Ghana (Rasul et al. 2017). While we address the same broad research questions linking the management practices bureaucrats operate under to bureaucratic performance, we do so while introducing some changes in the measurement of key variables. The first relates to measuring the output of bureaucracies. This is necessitated by the fact that civil service bureaucracies worldwide differ greatly in whether and how they collect data on their own performance. Unlike macroeconomic or household survey data, statistical agencies are typically not involved in measuring government effectiveness, and few international standards exist to aid cross-country comparisons.

In Ghana, we use quarterly and annual reports for each civil service organisation to build a database of 3,628 projects and tasks undertaken across 31 organisations during 2015. This allows us to build a comprehensive picture of what government bureaucracies actually do.

As Figure 1 shows, most of these projects are non-infrastructure projects, many of which are related to internal bureaucratic functioning rather than frontline public service delivery – the most common project type relates to human resource management (‘monitoring, training, and personnel management’). This reinforces the importance of understanding how the management practices that bureaucrats operate under relate to bureaucratic effectiveness.

Figure 1 What do bureaucracies do?

The second change from Nigeria to Ghana is an alternative approach to measuring management. While still anchored in the Bloom and Van Reenen (2007) and Bloom et al. (2012) methods, our approach in Ghana was designed to probe the sensitivity of results to differences in how such data are collected, and conceptual differences in the measurement of specific practices.

Our findings

Analysing this rich performance and management data together led us to three main insights that hold for both the Nigerian and Ghanaian bureaucracies.

1. There is considerable variation in performance across bureaucratic organisations within each country.

Figure 2 shows the variation in project completion rates across organisations in each country. There is huge variation in the performance of different government bureaucracies in each country. This variation occurs despite the fact that multiple organisations engage in similar project types, hires are assigned from the same pool of incoming bureaucrats, and many are located geographically close to one another in Abuja/Accra.1

Figure 2 Bureaucratic performance by organisation

2. There is huge variation in management practices across organisations.

In both Ghana and Nigeria, we focus on two dimensions of management practice: those capturing bureaucrats’ autonomy/flexibility, and those capturing incentives and monitoring for bureaucrats.2 For the autonomy index, we assume greater autonomy corresponds to better management practices (and similarly for the incentives/monitoring measure).

Figure 3 Management practices across bureaucracies

Figure 3 shows that, as with bureaucratic performance, there is high variation in practices across organisations. Again, this variation occurs despite the fact that all organisations in each country share the same colonial and post-colonial histories, are governed by the same civil service laws and regulations, are overseen by the same supervising authorities, are assigned new hires from the same pool of incoming bureaucrats each year, and are often located in close proximity to one another.

3. Autonomy is positively associated with bureaucratic output, but monitoring/incentives are negatively associated with bureaucratic output.

Since each organisation implements multiple project types, we are able to compare the relationship between management practices and output across organisations, holding constant the type of projects that each organisation implements.

In both Ghana and Nigeria, we find that:

  • the autonomy index of management practices is robustly positively correlated with project initiation, full completion, and average completion rate;
  • the incentives/monitoring index is robustly negatively correlated with all these measures of project completion.

Overall, in both settings, management practices for bureaucrats matter and are of economic significance. In terms of policy implications, the positive correlation of management practices related to autonomy with project completion rates supports the notion that bureaucracies could delegate some decision making to civil servants, relying on their professionalism and resolve to deliver public services. The evidence is less supportive of the notion that when bureaucrats have more agency, they are more likely to pursue their own, potentially corrupt, objectives.

The negative partial correlation between project completion rates and management practices related to the provision of incentives and monitoring of bureaucrats, is surprising and counter to evidence from private sector settings. Evidence on the impacts of performance-related incentives in public sector settings is mixed (often focusing on the impacts of specific compensation schemes to frontline workers). Ours is among the first evidence to suggest the possibility that such management practices negatively correlate to the output of the vital tier of civil service bureaucrats in multiple contexts. As such they serve as a useful check on the ‘New Public Management’ agenda sweeping through government bureaucracies, where the use of performance incentives is seen as a central aspect of updating bureaucracies.


Given the growing recognition of the role that bureaucrats and bureaucracies play in determining state capability, it will be important for researchers to understand similarities and differences across such state organisations in order to advance the literature. Our analyses contribute to the understanding of within-country variation in effectiveness and, further, highlight the role that management plays in driving pockets of good governance within similarly structured political institutions in relatively weak states. We hope our work will be the first of many to help establish a picture of bureaucratic effectiveness in different settings and the sources of within-country heterogeneity.


Bloom, N and J Van Reenen (2007), "Measuring and explaining management practices across firms and countries", Quarterly Journal of Economics 122: 1351-408.

Bloom, N, R Sadun and J Van Reenen (2012), "The organization of firms across countries", Quarterly Journal of Economics 127: 1663-705.

Finan, F, B A Olken and R Pande (2015), "The personnel economics of the state", in Handbook of Field Experiments, forthcoming.

Rasul, I and D Rogger (2016), "Management of bureaucrats and public service delivery: Evidence from the Nigerian Civil Service", Economic Journal, forthcoming.

Rasul, I, D Rogger and M Williams (2017), “Management and bureaucratic effectiveness: A scientific replication”, mimeo, UCL.


[1] As we use the minimum and maximum score of reports for the extensive margin of project output, it is possible that the percentage of initiated projects is below that for completed projects, as occurs in one organisation.

[2] Each index is converted into a z-score (so are continuous variables with mean zero and variance one by construction), where both are increasing in the commonly understood notion of ‘better management’.

Back to top

The Lives and Times of Civil Servants in the Developing World | VoxDev
25th August 2017

The characteristics of a nation’s institutions have long been regarded as fundamental to national development. Appropriately designed public institutions are increasingly seen as key to prosperity (North 1990, Finer 1999, Acemoglu et al. 2005, World Bank 2017). But who are the people that work in these institutions? What is their experience of being a public official in a developing country? In recent research, I brought together surveys of civil servants – the professional body of administrators who manage government policy – from across the developing world to provide micro-evidence on life inside government (Rogger 2017).

The organisations and individuals that make up public institutions have recently been the subject of increasingly rigorous empirical investigation (Iyer and Mani 2012, Dal Bo et al. 2013, Bertrand et al. 2015, Finan et al. forthcoming). So far there is not, however, a broad or ‘thick’ description of these organisations and officials based on rigorous data. Recent efforts to survey a large number of civil servants in a few bureaucracies have provided an opportunity to start building a picture of the ways in which officials in poor countries work. My research has combined direct surveys of 13,591 civil servants and their 204 organisations from Ethiopia, Ghana, Indonesia, Nigeria, Pakistan and the Philippines, to highlight the key aspects of what they are like, and their experience of civil service life.

From this, I can draw out five ‘stylised facts’ of the civil service in the developing world:

1. The civil service is a highly varied place

In contrast to the stereotype that the public service is unified in its routine, the surveys in every country show it to be diverse in key characteristics of public administration, from access to physical capital to the extent of political interference in the daily work of the service. While some local governments have regular internet access and computers for most of the staff, others don’t even have electricity. In some countries politicians rarely disturb their bureaucratic colleagues, but in other countries political interference is common.

Management practices in Nigeria illustrate this. I followed the World Management Survey methodology [http://worldmanagementsurvey.org/] to measure the quality of management at 94 organisations across the three tiers of Nigeria’s government. Figure 1 is a scatterplot of the aggregate scores across these organisations, showing real differences in management activities. Organisations at the bottom monitor their performance only when a new President is sworn in, while those at the top do so on a weekly or daily basis.

Figure 1 World Management Survey Scores across Nigeria’s government

2. The experience of the civil service is highly localised

Variation in the civil service is often apparent within small neighbourhoods of a few public officials or organisations. Civil servants within the same division may use different ways to further their careers. Some rely on their performance to move them up, and others ‘hustle’ for positions through their networks.

Returning to Figure 1, we can see the local nature of variation in management. The triangles represent local governments that were surveyed. The green filled triangles represent local governments in Kaduna state, oversampled to assess the extent of variation in the public sector of a single state. We see that Kaduna’s local governments range from the 90th to the 10th percentile of management quality in the country.

This diversity is represented in other features of Kaduna’s local governments. The number of hours of electricity available during a typical working day goes from zero in one local government to continuously available in another. Similarly, the percentage of officers with access to a computer varies almost uniformly throughout the distribution from 0% to 60%. Half the organisations have zero access to the internet, and the other half at least 15 hours a day. This illustrates the degree of heterogeneity within a single state.

3. The organisation plays a critical role in the experience of the civil servant

The vast majority of public officials have a long association with their organisation. The average civil servants in our data enter the service in their mid-twenties, spend a couple of years in an organisation, and then transfer to the organisation in which they spend more than a decade, if not more. Only a small minority of civil servants migrate across organisations in the service. This stagnation contrasts to the variety of posts held by AIS officers or other high-flyers.

Although federal ministries are literally across the road from each other in many cases, the ministry an official ends up in plays a substantial role in the quality of his or her working life. The experience of working at an organisation in the bottom quartile of management quality in Figure 1 would be very different to that of working at one what was in the top quartile. The surveys also imply that the organisation is a key determinant of job satisfaction.

4. The stock of human capital within the service is highly persistent

Civil servants enjoy employment protection, to guard them from political interference. Consistent with these protections, civil servants spend much of their working life in the civil service. As Figure 2 describes, the mean civil servant has spent three-quarters of his or her working life in the civil service, and 65% in the same organisation.

Figure 2 Characteristics of civil servants across countries

5. Non-market forces determine the incentives of civil servants

Though pay and performance-linked bonuses may be a factor in civil servant performance, other forces create incentives within the bureaucracy. Comparing across different questions on pay, we could correlate the extent to which satisfaction with the public sector is driven by satisfaction with wages. Within country, the correlation is surprisingly low: 0.20 in Ghana, 0.18 in Indonesia, 0.24 in Nigeria and 0.05 in Pakistan. These are large enough to be significant (positive) predictors of overall job satisfaction. This supports evidence from experiments such as Dal Bo et al. (2013) and Ashraf et al. (2015) that suggest high wages are an important part of the motivation to work in the public sector. However, this leaves the vast majority of job satisfaction unexplained.

The surveys imply that other factors are more important. For example, the extent to which civil servants believe they are working within a well-functioning organisation is more predictive of satisfaction than income in all of the surveys.

'Thick' descriptions of the civil service from this first round of surveys of civil servants provide details of the context in which specific reforms occur. As we develop the next generation of reforms, surveys and experiments, we should aim to embed our insights in the wider context of the world that public officials inhabit.


Acemoglu, D, S Johnson and J A Robinson (2005), “Institutions as a Fundamental Cause of Long-Run Growth” Chapter 6 of Aghion, P And S N Durlauf (2005), Handbook of Economic Growth, Volume IA, Amsterdam: Elsevier.

Ashraf, N, O Bandiera and S Lee (2015), “Do-Gooders and Go-Getters: Career Incentives, Selection, and Performance in Public Service Delivery,” HBS Working Paper. 

Bertrand, M, R Burgess, A Chawla and G Xu (2015), “Determinants and Consequences of Bureaucrat Effectiveness: Evidence from the Indian Administrative Service”, unpublished Mimeo.

Dal Bo, E, F Finan and M A Rossi (2013), “Strengthening State Capabilities: The Role of Financial Incentives in the Call to Public Service,” Quarterly Journal of Economics 128: 1169-1218.

Finan, F, B Olken and R Pande (forthcoming), “The Personnel Economics of the State”, Handbook of Field Experiments.

Iyer, L And A Mani (2012), “Traveling Agents: Political Change and Bureaucratic Turnover in India,” Review of Economics and Statistics 94: 723-739.

North, D C (1990), Institutions, Institutional Change, and Economic Performance, New York: Cambridge University Press, New York.

Rogger, Daniel (2017), "Who Serves the Poor? Surveying Civil Servants in the Developing World," World Bank Policy Research Working Paper 8051.

World Bank (2017), World Development Report 2017: Governance and the Law, Washington: World Bank

Back to top


Water Get Enemy: A Graphic Novel on Governance | Governance for Development 
13th July 2017


This blog post is part of a series for the 'Bureaucracy Lab', a World Bank initiative to better understand the world's public officials.

“Why? Why do we always fail the people of this country?” So reflects the public official who plays the hero in my graphic novel on governance in the developing world. The story, set in fictional Zanzarim, follows the struggles of the ‘Director’ up to that point, as he labours to implement policy that will help his fellow citizens. His exhausting — and frequently unsuccessful — attempts to succeed mirror the many such struggles I have witnessed in the governments of developing countries across the world.

We live in a time when public trust in government is at historically low levels. Many citizens ask the same question as the Director – why does government fail to serve us? These concerns are particularly well-founded in the developing world where service delivery is often patchy or non-existent. 

The answer to these questions lies in the many complex levels of the policy chain. At each step, the Director must fight to ensure that good policy isn’t warped to help special interests, or weakened at the request of unresponsive or overburdened bureaucrats.  And at each step, the effectiveness of policy is chipped away. In the end, it is only because of the Director’s dogged determination that anything happens at all. He represents one of the heroes of development that populate public services.
After working in government in a range of countries, I began to wonder how I could immerse others into the world of officialdom that I experienced, so that they could meet those like the real-life Director.  One way is to paint a picture of officialdom using available statistics, as I do here. Another (perhaps a bit more engaging) way is to partner with a Nigerian comic book artist and write ‘Water Get Enemy’, a graphic novel illustrating governance in the developing world.
The more I talked to others in service around the world, the more I realized how similar our experiences were. And the more time I spent in service, the more affected I was by the common saga unfolding around me. To share that story with the world in a visceral way, I created ‘Water Get Enemy,’ with graphic artist Albert Ohams.
“Isn’t a graphic novel on governance where street cred goes to die?” asked a colleague of mine. Perhaps, but it also provides a window into the battles that must be waged inside the bureaucracy to deliver public services, and the threats that cause government failure.
So put yourself in the Director’s shoes, and walk into his steaming office (the electricity is down, so the AC is off). Follow him through days brimming with absent statistics, episodes of political interference, gross bargains for funding, and the failures of the private sector and communities that buffer those of the public sector. Feel his frustration and relief at the world in which he must work. When the Director’s Permanent Secretary asks him to change the location of some public wells so that they are in his friends’ communities, what should he do?

In ‘Water Get Enemy,’ the Director’s world is a difficult one. But the story is not intended to bemoan the state of the world and its dysfunctional bureaucracies. Rather, it’s meant to highlight that there are passionate and competent people in governments throughout the developing world. They may be continuously challenged by the environments in which they work, but they still get done what has to be done. They change the way these governments work from the inside out. The story is thus a salute to battle-hardened public officials; the real-life heroes of development.
To the numerous heroes and heroines I’ve worked with, here’s one attempt at illustrating your struggles and many successes. Though mine, at the very end of the book is not my best look, I hope you like your cartoon likeness.
‘Water Get Enemy’ (named after the Fela Kuti song of similar name) is freely-available online at www.watergetenemy.org

Back to top


What type of bureaucrat are you? | Governance for Development 
22nd May 2017

This blog post is part of a series for the 'Bureaucracy Lab', a World Bank initiative to better understand the world's public officials.

In the world of public sector bureaucracy, what type of bureaucrat are you? 
In the board game 'Bureaucracy', you must assume the role of the ‘Lifer’, the ‘Over Achiever’, the ‘Empire Builder’, or the ‘Hustler’.  Each character must use different tactics associated with their personality to rise up the ranks of the bureaucracy to achieve the position of director.  For example, by amassing contacts, the Hustler can attempt a 'power play' on players above her in the hierarchy. 

Though a rather tongue in cheek look at the everyday lives of so many of us working in large organizations, this board game has given me a different lens through which to understand the world of the public sector bureaucracy. 
In a new World Bank Working Paper, 'Who Serves the Poor?', I use eight surveys of civil servants from six countries - Ethiopia, Ghana, Indonesia, Nigeria, Pakistan and the Philippines - to explore the characteristics and experience of life as a public official in the developing world.  Evidence from these surveys implies that the board game isn’t too far from reality.
So what kind of bureaucrats are out there?  Across the civil service, ‘Lifer’ seems the dominant personality type.  In almost all of the countries surveyed, officials entered the service when they were in their mid-twenties, worked for a couple of years before joining their current organization and stay there for a decade or more. But staying doesn’t necessarily mean they are satisfied.  Officials seem to get stuck, with half of those interviewed in Nigeria wanting to have been moved more. A dominant reason for promotion across the surveys is simply that the official has been around long enough, and it’s her turn for promotion.
That's not to say there isn't room for some ‘Over Achievers’.  In Ghana, in particular, two-thirds of officials believe that ability is an important predictor of promotion.  I found evidence consistent with the idea that ability is one route to success in the public sector, which echoes previous research.  However, what it means to be an overachiever varies across settings.  The surveys imply that there is a quite a bit of variation in the average level of education across and within civil services.
The other route to success would seem to be hustling, criss-crossing the civil service in an ascent to the top.  This small minority of such officials is more likely to state that they had control over their career progression and that they had had ‘influence’ on securing their current posting.  These officials, as in the board game, have more of a network, and use it to their advantage.
Hardest to identify are the ‘Empire Builders’.  This might partly be because empires arise opportunistically.  As one spot on the board game puts it, 'Crisis! Payoff: 3 staff, 3 civil service ratings'.  One strategy is to proxy empire by the amount of time officials spend at work beyond their contracted hours, assuming that empires take investment.  Though around 17% of staff are said to be working less than their contracted hours, a third work more, perhaps looking for their moment to empire build.
The surveys I study imply that these personality types are quite consistent across services, echoing some of the surprising similarities across bureaucracies.  Perhaps it is these commonalities of bureaucracy that make it possible to have a board game whose rules are a 'Code of Bureaucratic Regulations' or 5 'stylized facts' of the civil service that I conclude the paper with.  But I also find that the implied mix of personalities varies a lot, with lots of politics, favoritism and hustle in Nigeria and Pakistan contrasted with relatively frequent isolationism in the Philippines.
It’s crucial to understand the personalities that make up the developing world's public sectors.  They are instrumental in ensuring the world's poorest people have access to public services.  Understanding "the glamor and excitement of civil service inaction," as the 'Bureaucracy' board game tagline puts it, is central to securing services for the world’s poorest.

Back to top