Public service reform

The focus in this Pack is on action: on the public service reform intervention, conceived as an exercise in helping. The readings included in this pack have been chosen to be interesting and fresh rather than comprehensive; that is, to stimulate thinking rather than necessarily to ‘cover all the bases’ in public service reform as currently practised. For a more comprehensive review of debates and evidence on this subject, please see the GSDRC’s Civil Service Reform topic guide.

Reading 1 (McCourt) is a map of the territory, which covers a bigger area than is often realised. It also introduces the problem-solving approach, which is developed in Reading 2 (Andrews et al.), and the political and institutional dimensions of reform which are further developed in Readings 3 (Bunse and Fritz) and 4 (Tavakoli et al.). The problem-solving approach is visible in all of these readings. For example Bunse and Fritz draw on Kingdon’s seminal 1995 book on the policy process to make the point that policymaking starts with the identification of problems, to which policymakers then propose solutions in the form of policies which direct political actors’ attention to those problems.

Reading 5 (Halleröd et al.) further develops the application to service delivery, showing that service delivery is affected by what they call ‘quality of government’. It is also, in a way, a bridge between the first four theoretical readings and our more practical final readings, which take us all the way down to the organization level of reform. In Reading 6, Leonard grapples with what makes it possible for a single organization, including a government agency, to be a ‘pocket’ of effectiveness in states that are predominantly weak. Readings 7 and 8 (Nunberg and Taliercio, and Fujita) represent examples of two such ‘pockets’.

What is the universe of public service reform? Why is it important for the practitioner to have a range of reform models at his or her fingertips?

Reading 1: McCourt, W. (2013) Models of public service reform: A problem-solving approach, Washington DC: World Bank, Policy Research Working Paper No. 6428
See revised post-publication version, retitled: Towards ‘cognitively complex problem-solving: six models of public service reform

This paper includes a new argument about the value to practitioners of having a range of models at their disposal as an aid to reform problem-solving. It introduces the currently fashionable problem-solving approach, and also the political dimension of reform, both of which are developed in later readings.

Table 1 in the Reading distils much of its message:

Public Service Reform Problems and Models
Problem Model Main Action Period
How can we put government on an orderly and efficient footing? ‘Weberian’ public administration and capacity-building Post-independence period in south Asia and sub-Saharan Africa
How can we get government closer to the grassroots? Decentralization 1970s to present
How can we make government more affordable? Pay and employment reform 1980s and 1990s
How can we make government perform better and deliver on our key objectives? New Public Management 1990s to present
How can we make government more honest? Integrity and anti-corruption reforms 1990s to present
How can we make government more responsive to citizens? ‘Bottom-up’ reforms Late 1990s to present

What is the potential of the problem-solving approach? What are its pitfalls? Can it be taken to scale?

Reading 2: Andrews, M., Pritchett, L. and Woolcock, M. (2013) ‘Escaping capability traps through problem driven iterative adaptation (PDIA), World Development, 51: 234-44.
Ungated version: CGDEV Working Paper 299, 2012

A number of writers have converged on a problem-solving approach which focuses on solving locally-defined problems through experimentation and rapid learning. Andrews et al. have provided the fullest exposition to date.

Is political and institutional analysis an exercise in crying over spilt milk? It explains, but can it predict? How should political analysis inform the design of public service reform interventions? What possibilities for donors follow – and, likewise, what limitations on what donors can do?

Reading 3: Bunse, S. and Fritz, V. (2012) Making public sector reforms work: Political and economic contexts, incentives and strategies, Policy Research Working Paper 6174. Washington, DC: World Bank

Reading 4: Tavakoli, H., Simson, R. and Tilley, H. with D. Booth (2013)Unblocking results: Using aid to address governance constraints in public service delivery, London: ODI
See 8-page briefing note based on this synthesis paper.

In a way, the 2000s were the decade of politics, with investments in political economic analysis (PEA) by DFID, SIDA, USAID and the World Bank, building on studies from the 80s and 90s of the politics of structural adjustment by writers like Nelson, Whitehead and Williamson. However, the complaint has arisen among practitioners that it is impossible to read off programming choices from the analyses. Bunse and Fritz’s paper responds to the complaint. They highlight the need to differentiate between country contexts at the project preparation stage, concentrate on reform implementation when temporary windows of opportunity open up, and design reforms which take account of the stakeholder incentives which PEA identifies.

Tavakoli et al.’s paper develops Bunse and Fritz’s work. It brings in the institutional dimension (that is, the framework of rules, traditions and organizational structures which shapes the way that development actors behave), and it appapplies PEA specifically to public services. It also puts international development agencies at the centre of the stage. (On the donor role, see also Reading 7.)

What evidence do we have that public service reform actually improves important development outcomes? What evidence would we like to see?

Reading 5: Halleröd, B., Rothstein, B. Daoud, A. and Nandy, S. (2013) ‘Bad governance and poor children: A comparative analysis of government efficiency and severe child deprivation in 68 low- and middle-income countries’, World Development, 48: 19–31
Ungated version: QOG working paper 2012:9

Reading 5 is the most rigorous empirical study among the readings in our pack, and includes a discussion of strengths and weaknesses in the available evidence about the effect of public service reform on development outcomes. It introduces the important work of the Quality of Government Institute at the University of Gothenburg, which is not as well known as it should be outside Scandinavia.

Can we reform anything without reforming everything?

Reading 6: Leonard, D. (2010) ‘“Pockets” of effective agencies in weak governance states: Where are they likely and why does it matter?’, Public Administration and Development, 30: 91-101
Ungated version: IDS Working Paper 306, 2008

Reading 7: Nunberg, B. and R. Taliercio (2012) ‘Sabotaging civil service reform in aid-dependent countries: Are donors to blame?’,World Development, 40: 1970–81

Reading 8: Fujita, Y. (2011) What Makes the Bangladesh Local Government Engineering Department (LGED) So Effective? Working Paper 27, Japan International Cooperation Agency Research Institute.

Public service reform is staggeringly complex: so many political and institutional ducks to line up, and at so many levels, from the capital city to the remotest outpost of the State. Most public servants, even very senior ones, have little access to the big levers of government-wide reform. Similarly, international development agencies find themselves interacting for much of the time with individual public agencies. If they are reform-minded, public servants and development agencies will want to start from where they are, namely in an individual agency, rather than from where they might like to be, namely at the level of government as a whole.

Reading 6 is a thoroughgoing review of the experience of ‘reform in one agency’, and the conditions which need to be present for reform at that level to succeed. Readings 7 and 8 are examples from Cambodia and Bangladesh of ‘reform in one agency’. As it happens, the Cambodian reform didn’t stick, while the Bangladesh reform did, at least for a decent period. However, taken together, they shed light on the real-world tactical question of the level at which a reform intervention can and should take place.

What should our working model of public service reform be now? What is the best way for us to help governments to solve their problems?

For most of the 2000s, a decade which opened with the international commitment to the MDGs, observers of what passed for debate in public service reform would have been forgiven for quoting the words of Albert Einstein in a different context: ‘Everything has changed except our way of thinking.’ Developing country reality and the development debate had moved on, but public service reform was still stuck in 1990s thinking. Of course, the practical work of public service reform must remain the spasmodic business that it has always been, elegant laboratory models notwithstanding. However, in the readings we see, arguably, the dim outline of a coherent new approach.

It is an approach which starts from where governments are, with the problems they are trying to solve, rather than starting with the latest reform ‘widget’ (Anuradha Joshi and Peter Houtzager’s useful word), and forcing the square peg of government into the widget’s round hole.

However, the approach also recognizes that in a globalized world, governments don’t reinvent the wheel, but draw on current international models. Problem-solving does not abolish ‘best practice’. On the contrary, it makes knowledge of a broad range of models indispensable.

The new approach recognizes, further, that in terms of many governments’ current priorities, public service reform is not an end in itself, but a means to other, larger ends, notably major development outcomes in which improved public services are instrumental. (That is why we have preferred ‘public service reform’ as our label, rather than ‘civil service reform’.)

The new approach is also opportunistic: it recognizes that the best should not be the enemy of the good. To be sure, comprehensive, whole of government reform remains the ideal. However, in the absence of such whole of government reform, the new approach gladly falls back on the reform of an individual public agency as a Next Best Thing, and as an incubator for system-wide reform later on.


Send feedback on this reading pack