GPSA Knowledge Platform


Political Economy Analysis – A step forward for social accountability and development, or three steps back?
Part II

Colm Allan

In a previous post, I argued that mainstream PEA has limited our understanding of how to achieve better governance and development outcomes. In part, this limitation arises from their Public Choice Theory (PCT) premises, which assume that all individuals and their decisions are motivated by incentives or expected payoffs; positive incentives reward those who comply with institutional rules and negative incentives punish those who don’t. What mainstream PEA offers, then, is a focus on how politicians, bureaucrats and civil society actors respond to incentives created by formal and informal institutional rules.

But what PEA does not offer is a systemic understanding of the processes that are required in order to make socially accountable governance institutions work in the first place. I argued that for institutions to be effective and for public policies to be successfully implemented in ways that address priority needs and rights, requires public resource management processes.

Even if you subscribe an analysis that focuses on individual and group self-interests, the only way in which politicians can ensure the provision of public goods and services that meet the interests of “special interest groups” is through the manipulation of these public resource management processes. For this reason, even from the point of view of mainstream PEA you still need to explain what happens within these processes in order to understand development challenges. You need to be able to investigate and explain capacities, as well as contestation, within each process. This includes: how bargaining between competing interest groups is conducted through needs assessment and policy making processes; how priorities are negotiated within strategic planning processes; how special interests are met by resource mobilisation (including revenue collection) plans; how resource are allocation (budgeting) ends up serving special interests; how revenue collection and expenditure and performance are managed in ways that serve selected interests (including those of bureaucrats); who benefits from the way corrective and preventative action are taken or not taken; and finally, how auditing and oversight are undertaken and whether this reproduces or challenges vested interests.

Each of these processes (or sub-processes) represents a site of engagement between competing interest groups seeking not only to maximise their welfare, but also to realise their concerns and needs. Each is a site of contestation and negotiation; whether by non-decision making, the exclusion of potential decisions, or the active and transparent domination of decision taking by powerful interest groups. Without an understanding of the role played by these processes we are left with a model of governance and development that is made up of institutional rules on the one hand and self-interested economic individuals and interest groups on the other; with nothing in between. We are left with no insights into the sites at which this contestation over public resources takes place.

PEA, following PCT, assumes that if governance institutions aren’t working it must be because people lack the necessary incentives and payoffs to comply with institutional rules.
But institutions do not simply consist of rules. They are made up of roles. And the ability of institutional actors to perform their roles in a socially accountable fashion is not simply a matter of incentives. It is a matter of having the capacity to explain and justify the use of resources to meet priority needs (by supply actors), to engage with the evidence base for these explanations and justifications (by demand actors) and to ensure corrective action where they don’t.

We need a framework for analysing the relationship between these capacities, processes and institutions. We need to test theories of change which offer tools for building these capacities in ways that make processes more socially accountable and which enable institutional and policy effectiveness.

We also need a framework that can analyse the relationships between these processes and bureaucratic and governance institutional arrangements across tiers of governance. Do these institutional arrangements enable or constrain social accountability processes? Is the system integrated? Or does it breakdown between levels of government?

I don’t see how mainstream PEA and its game theory models bring us any closer to the development and testing of coherent theories of change for how to strengthen systems of socially accountable governance and development (including institutional arrangements, social accountability processes, and the capacities of state and non-state actors).

By contrast, I think that the process-based Social Accountability Monitoring and Evaluation (SAME) framework and indicator set provides a coherent and testable theory of change for evaluating development and governance challenges. It provides a basis for diagnosing capacity constraints, for analysing the influence of vested interests and for recommending programme interventions. The pilot application of this framework in the health sector in Mozambique in 2013/14 (supported by SDC) pointed to the way in which powerful interest groups had influenced the degree of social accountability of key processes and limited institutional effectiveness.

For one thing, the ruling party insistence that its electoral manifesto constituted the equivalent to a needs assessment certainly contributed to the failure to undertake a rigorous health needs assessment process. Secondly, donor insistence on predominantly epidemiological indicators in the Mozambican health SWAP had the unintended consequence of a curative emphasis within expenditure reporting, performance reporting and oversight processes.

But, even if these political “incentive structures” were to change (with the ruling party giving approval to rigorous bottom-up needs assessment and health planning processes, and if donors balanced the inclusion of preventative and curative performance indicators within the health SWAP, this would not change existing process capacities. Nor would it address the lack of systemic integration of processes across tiers of governance.

For instance, there was no capacity of supply actors to include narrative explanations of their performance at any of the 7 health sector pilot sites (which spanned all 4 tiers of governance). This placed obvious limits on the ability of demand actors (CSOs) and oversight bodies to engage with performance and advocate for corrective action. Similarly, the lack of capacity to undertake programme based budgeting, together with this absence of programme based performance reporting, had a negative effect on the capacity of the health sector (across tiers of governance) to engage in effective risk management. This in turn not only undermined corrective action but weakened oversight processes.

The point is that the capacities of actors within each of these processes places real limitations on the implementation of health policy in Mozambique. What the SAME framework shows is that ultimately these supply and demand capacities, and the relative powers of vested interest groups within the state and society, along with the influence of institutional rules and roles, converge to form an emergent power that enables or constrains socially accountable policy outcomes. These outcomes can’t be predicted or explained by the self-interested pursuit of payoffs by individual policy makers and actors or by using the limited tools of mainstream PEA.

About the Author

Colm Allan

Colm Allan is a research associate at Rhodes University, South Africa, and an independent consultant specialising in social accountability monitoring and evaluation. He has been involved in applied social accountability monitoring work for the past 17 years, founding the Public Service Accountability Monitor in 1999 and the Centre for Social Accountability in 2006 (both at Rhodes University). He conceptualised and developed a five-process based approach to social accountability, which has subsequently been adopted by numerous civil society organisations engaging in social accountability monitoring and advocacy across Southern Africa. Since 2012 he has been developing a set of process-based social accountability monitoring and evaluation (SAME) indicators. These were piloted in Mozambique in 2013/2014.

3 Responses on Political Economy Analysis – A step forward for social accountability and development, or three steps back?
Part II"

  1. Varja says:

    Thanks Colm for the contribution. I am undecided on your point (on how many steps forward and backward PEA brings us) — mostly because I think a tool such as PEA can be used in very thoughtful ways. We have used it ourselves at Twaweza – not feeling at all locked into the theoretical underpinning of PEA, but using it as a springboard (and a way to discipline our thinking) for looking not only at incentives and formal rules of the game, but capacities, informal rules of the game (motivations to change, or motivations – i.e. barriers – against change), as well as relationships between the various actors. However — I am always interested in new, different tools which might do similar, but better — so do you have a link to the SAME framework you referred to? Both the tool itself, as well as the process and result of its application in Mozambique? And, last but not least, are there reflections in terms of how much better the Mozambique health sector analysis fared as a result of this tool? Thanks!

  2. Colm says:

    Dear Varga

    Thanks for your very insightful comments. To be a bit clearer, I am not suggesting that PEA cannot ever be usefully employed. At a micro level of analysis, I think it can often be very insightful in explaining the motivations of individual actors. Where I am more sceptical is my belief that the standard micro-economic game theory explanations of “rent seeking”, “special interest groups”, “clientelism”, etc, within PEA provide little help in explaining systemic governance and development issues.

    I think it is important for social activists and development practitioners alike to be able to say something meaningful about the progress we are making in strengthening governance and development at a systemic level. For instance, and very practically, what can we say about the $2.4 billion worth of OECD aid for governance and civil society interventions in Sub-Saharan Africa in the year 2014? When I ask this question I don’t see any compelling big picture answers. What I see is fragmentation. Lots of micro-level insights from many different and often isolated interventions, even within single policy sectors in the same country.

    What’s missing is some way of being able to connect up the multiple micro-level interventions by social activists, state actors and development practitioners (donors). What impact are they having on strengthening governance institutions and in implementing policy undertakings to provide public goods and services that meet priority social needs.

    I think that this gap needs to be filled by focusing on processes. Development policy has been enveloped in mysticism for the past 30 years; it has argued as an article of faith that institutions are necessary for effective governance. But it doesn’t explain how good institutions will convert their policies into services and public goods that will meet social needs. This is left to the realm of magic. With New Institutional Economists in the 1980’s and 90’s the assumption was that developing country governments simply need to adopt the right “rules of the game” to ensure economic development. Problem solved. But when the problem wasn’t solved, along came Public Choice Theory with its contention that the reason why the “rules of the game” have not been implemented is because state and non-state actors lack the necessary incentives. This is where PEA enters the picture. After the 2000’s the focus has shifted to sifting through the murky underworld of politics to identify the incentives of key political actors in trying to implement or to block these rules.

    Whilst this is all very engrossing and stimulating for those who love solving “whodunit mysteries”, and for those who, like myself, suffer from ADD, it doesn’t explain whether we are making systemic progress in realising (ever changing) social needs.

    But by looking at public resource management processes, and specifically by looking not only at how effective but how socially accountable these processes are, we can actually build a bridge between the level of governance institutions and public policies and the practical realisation of people’s needs and concerns. I am proposing that by looking at:
    • the ideas,
    • the practices (behaviour) and
    • capacities
    of state and non-state actors within each process, and the engagement between these actors, we can build a systemic picture that explains actual service delivery and human development outcomes.

    Ultimately, social accountability is about generating explanations and justifications for how available resources are used to meet priority social needs, and it is about taking corrective action when priority needs are not met as effectively as possible.

    The idea behind the set of social accountability monitoring and evaluation (SAME) indicators that I have been working on for the past couple of years, is find a way of drawing up a baseline for what the current ideas, behaviour and capacities of state and non-state actors is within each PRM process. Do they accept the idea that they should produce explanations and justifications (state actors), or that they should interrogate the evidence-base for these (non-state actors), and engage candidly around these (state and non-state actors) and take corrective action where needs aren’t effectively met (state actors)?

    Do they engage in the actions, behaviours and practices consistent with the idea of social accountability? And importantly, do they have the capacity to give effect to a socially accountable PRM process? This is where politics and power analysis are critically important. Do state actors and social activists have the necessary resources (material and ideational) to build and maintain this capacity.

    Using this baseline information, we can then develop theories of change and design contextually relevant interventions to try and strengthen these processes. We can build M&E frameworks into our project design to track our contributions to changing current ideas, practices and capacities. And using these indicators we can also conduct external evaluations of the impact of our interventions. In turn, if we learn from multiple evaluations we can start building a systemic picture of social accountability in a particular policy sector and particular country.

    So power analysis, including the distribution of resources, forms a key part of the SAME framework and indicator set that I am working on. But for the time being I would prefer to call this power analysis rather than PEA to reflect its systemic and needs-based, rather than micro-economic, focus.

  3. Colm says:

    Oops. Apologies for misspelling your name, Varja.

Leave a Message

Your email address will not be published.

How Can I Contribute to the Knowledge Platform

You may contribute to the Knowledge Platform in many different ways: you can send and disseminate your social accountability materials (toolkits, reports, videos, etc.) in the knowledge repository; you can contact, interact and collaborate with other peers and join a global community of social accountability practitioners; you can participate in the different learning and knowledge exchange activities of the GPSA KP such as online courses, thematic forums, webinars and blogs; and you can develop a partnership with the GPSA KP to implement collaborative knowledge activities.

2016 GPSA Knowledge Portal. All rights reserved | Terms and Conditions
To participate in all the learning and sharing activities, you need to be registered Click here to create your account