Learning about the ‘mechanics of governance’ with DFID colleagues & researchers

This blog was originally published on the DFID Research blog and was called ‘The machinery of government and the mechanics of governance: Findings from the Uganda Governance Evidence Week‘. It was co-authored with DFID colleagues – Peter Evans, Alisha Patel and David Pedley.

In October 2018, DFID Research and Evidence Division (RED) and DFID Uganda piloted a new approach to research engagement and uptake and to cross-portfolio learning — the Uganda Governance Evidence Week.

This blog summarises key themes that emerged and what we learnt. It tries to capture illuminating discussions of the dynamics within governance sub-systems; the linkages between sub-systems; and how these together inform our understanding of governance and politics in Uganda.

While discussions centred on governance and politics in Uganda, wider themes of informality, ‘pockets of effectiveness’, unintended consequences of reform and adopting a multi-disciplinary and political settlements approach are relevant to many countries in which DFID operates — particularly our understanding of the sorts of conditions under which positive reforms can occur.

What do you think? In writing this blog, we hope to receive feedback and reaction to our governance research portfolio and this pilot approach to research engagement.

Designing Uganda Governance Evidence Week

The iterative process of the evidence week brought together DFID-funded researchers working on governance and politics in Uganda to discuss research findings and to reveal potential commonalities. We organised a series of events and workshops to facilitate engagement between DFID and UK government staff, researchers, Ugandan practitioners and policymakers, civil society organisations and donors to consider the policy and programmatic implications of the research we commission and fund. The week enabled us to think comparatively and systematically about governance and politics in Uganda, to deepen our cross-portfolio learning and to discuss future research priorities.

Breakout sessions for the external workshop were organised across 4 sub-systems, with researchers from different DFID-funded programmes engaging with Ugandan practitioners and policymakers, donor agencies, civil society organisations, and representatives from the private sector.

We organised and conceptualised the week with two ordering themes in mind:

i. ‘The machinery of government and the mechanics of governance’: the metaphorical device of the ‘machine’ is a longstanding and recognisable rhetorical tool in political science, and was used to facilitate discussions about systems, relationships and interdependencies. We aimed to get out of sector or research silos and help participants examine the whole complex system of governance, including sub-systems (broadly defined as the systems e.g. policy and delivery frameworks and networks and partnerships between government ministries and agents that lie within the core functions of the state). This also includes consideration of legislation, reform, feedback loops, interdependencies and the actors (the ‘mechanics’) involved. The machine metaphor also referred to the skills that actors (the ‘mechanics’) need in order to diagnose and repair the ineffective machine system and its sub-systems (both technical expertise and understanding of the political context and actors). This does not necessarily reflect how DFID sees its approach to governance research and programming.

ii. ‘The political and technical aspects of the “governance machine’: the focus on technical and political aspects stands as both a critique of ‘traditional’ approaches that focus exclusively on technical aspects, and equally a critique of newer governance approaches that focus exclusively on political aspects and underplay (or ignore) the technical challenges that remain to both government and donor programmes. Everything is simultaneously technical and political. During the workshop, we stressed the importance of looking through both ‘technical’ and ‘political’ lenses to understand inefficiencies, institutions and incentives as well as ideas, norms, values, networks and behaviours.

How complex and overlapping systems and networks are shaped by formal and informal dimensions

Informality and weak institutions often result in ‘functional deviance’.

Functional deviance can be defined as deviant practices that do not align with formal rules (e.g. payment of bribes) but are nevertheless functional because they manage against the odds to solve problems in a broken system. Such problems include accessing services in a context of material and human resource scarcities. Other problems are bigger, such as finding ways to fund election campaigns when formal funding mechanisms are non-existent and winning government contracts when bidding systems are weak. Informal practices ‘get things done’ for some but almost always at the expense of some others. In a sense they perform a “particularistic” governance function because of their reliance on personal relationships and the norms that bind these networks; yet this compromises the performance of formal institutions that are meant to function on the basis of “universalistic” or impersonal criteria.

Imposing ‘blueprint’ or generic prescriptions without accounting for the particularities of the context will, in all likelihood, be ineffective, especially if the functionality of informal practices is not taken into account in devising solutions to improve formal governance functions.

Informal networks can be bad, good, or both simultaneously, but informal governance is hard to escape.

Co-opting individuals, groups and networks continues to play an important role in Uganda under the National Resistance Movement (NRM), the ruling political party since 1986. As a result, informal networks shape the distribution of power, the structure and leadership of government organisations and relationships between public and private sectors. Informal networks don’t necessarily lead to negative outcomes, and research shows that they can promote progressive behaviours, integrity and professionalism (as demonstrated in the Ministry of Energy and Mineral Development), but it can simultaneously embed nepotism, corruption and inefficiency. Informal networks aren’t always simply ‘good’ or ‘bad’; they can be instrumental for some, even if not for others who are excluded from those networks or who, for instance, are precluded from certain services at the expense of the privileged groups. Importantly, belonging to these informal networks offers a promise of a better future — particularly as these networks can fill the space left by the unfulfilled promise of development.

Informal networks are a type of social bond and commonly reinforce social norms of solidarity and reciprocity. They are lubricated by trust and loyalty; this makes them powerful and difficult to dismantle or exit given the social and economic consequences of deviating from the network’s norms. Disrupting informal networks can also have negative consequences, disrupting ‘deviant functions’ and viable ways of getting things done, as well as the positive ones we usually assume. Informal networks may prevent systems achieving their full potential but may simultaneously prevent total collapse and failure.

Until we better understand how these often hidden (particularly to outsiders) networks and behaviours work, incrementally shifting informal networks, incentives, and behaviour out of the darkness (illicit, illegal) and ‘towards the light’ (legitimate, legal) could be a more effective, and politically viable, reform strategy. However, critics may oppose a ‘nudging’ approach in favour of more visible and punitive enforcement. This is a conundrum: which is preferable — a morally righteous strategy that does not work, or a morally ambiguous approach that does? And, importantly, given the potential for negative consequences of reforms — who decides?

Those not paying income tax may pay large amounts of ‘informal taxes’, such as bribes and ‘gifts’

These may allow or accelerate access to goods and services. Given the strength of networks and norms, and the importance to service users of obtaining access to these services (e.g. healthcare), it can be difficult to find entry points to stop these informal practices and substitute the informal social bonds with formal ones — the social contract between tax payer and state. Entrenching this social contract is likely to be a longer-term aspiration but is an important mechanism to support a range of progressive changes conducive to more inclusive democracy, stability, and a rules based system.

The evidence suggests that we should recognise that formal and informal systems, and formal and informal rules and relationships, operate simultaneously, and that we need to consider the effect of reforms in one area on the performance of the other. ‘Nudging towards the light’ could be more effective and politically viable than aggressive disruption, though this has not been rigorously tested and should be a priority for future research.

New data could transform our understanding of networks — and support more action

Opportunities include ‘data matching’ between different data sets — such as between income tax data and public procurement data — and also relating such data to more ethnographic and political research or data on networks, including the emotional impact of being in a network. This is worth exploring with research, though data quality may be a barrier, and researchers may not have access data beyond, perhaps at best, anonymised data.

How can change happen in Uganda?

Progressive reforms — even unexpected ones — are more likely to work when they align with a political agenda.

Examples offered include the Ugandan government’s need to focus on economic management as a political campaigning issue and to gain legitimacy amongst donors. This resulted in the Uganda Revenue Authority (URA) bringing more high net worth individuals (HNWIs) into the tax net and the Ministry of Finance (MOF) and Central Bank becoming bureaucratic pockets of relative effectiveness.

‘Pockets of effectiveness’ (POEs) are more likely in relatively simple sub systems with fewer moving parts.

Most POEs identified in the research happen in new policy and executive functions, particularly those linked to economic performance — e.g. the Ministry of Finance, Central Bank, oil management) and by their nature can be delivered by a few dynamic actors with specialist skills. This is not a coincidence.

Service delivery systems are by nature far more complex, involving more moving parts, functions, actors, interests and layers. Change requires a joined-up approach spanning local to national government, something that is a challenge in resource scare contexts. However, in Uganda, emerging research suggests that one service department — the National Water and Sewerage Corporation — appears to be a POE. This may be because many of its tasks are more functional/logistical than transactional, for example laying pipes rather than changing behaviours around education or health. Further research is required to understand how it performed so well relative to other service delivery ministries.

Reform may have unintended consequences that can undermine change or even cause a net deterioration in a wider system

For instance, top-down action by the newly established Health Monitoring Unit (HMU) to interrupt bribery in health facilities resulted in few bribes being paid, but it also resulted in lower staff morale and unprecedented nation-wide strikes, which is likely to prove unsustainable in the long term. It also disrupted the informal networks and ‘deviant functions’ that enable some service delivery to continue in a system under stress. We don’t yet know what the unintended consequences of this might be.

Informality, weak structures and systems, as well as limited accountability, mean that one targeted initiative to reduce corruption is unlikely to eradicate it. Instead, those benefitting from corruption are likely to shift their efforts elsewhere and find different entry points within the system. For example, research suggests that transparency reforms in public procurement displaced corruption elsewhere in the system. However, over the longer term regular displacement could increase the costs of engaging in regular and successive corrupt practices and reduce the returns on corrupt behaviour, but this needs to be tested. Furthermore, access to good quality data and enhanced transparency can help identify where corruption moves to when displaced.

Oversight mechanisms take many forms, with some more effective than others, and others working in some situations but not others.

Discussions over the course of the week on research programmes aiming to improve oversight in Uganda demonstrated that some mechanisms worked well: e.g. strong external professional networks, strong management leadership and independent opposition, as demonstrated in a randomised control trial of local politicians and bureaucrats in areas where the opposition had influence. However, heavy-handed and/or top down oversight, such as the case of the HMU, may have been effective in the short-term but is likely to be counterproductive and unsustainable in the longer term. Emerging evidence suggests that the effectiveness of interventions like transparency, accountability and community level monitoring vary according to the local political settlement that they are applied in, with impact greater where there is a real contest between ruling party and opposition at local level or developmental coalitions with the capacity and commitment to devise and enforce innovative approaches to governing the sector.

With thanks to our research partners for their comments and feedback on this piece.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s