This blog post is based on a presentation delivered by Dr Pallavi Roy on “Political economies of scaling fragile contexts” for the annual forum of Communities of Practice for Scaling. 

Scaling is the process of successfully expanding, adapting, and implementing policies or programmes to more beneficiaries over time with a specific impact in sight and requires mobilizing and coordinating a range of activities to achieve the desired impact. Successful scaling in fragile contexts has the added imperative of ensuring coping capacities are raised and the efforts are sustainable. Rather obviously this requires governance and enforcement capabilities that have to be commensurate with the desired objective.  

However governance is at a premium in fragile contexts. Fragility is a disparity between risks and coping capacities to address risks. In these fragile settings, SOAS-ACE’s research program has had to grapple with complex risks that were impediments to anti-corruption and development prospects, and therefore required solutions for resilience.  The fragility that we are referring to does not always mean conflict-affected but also vulnerability across all the dimensions of OECD-DAC’s framing of fragility index. 

Fragile contexts have especially weak(er) governance systems. This is because of rent capture and leakages that are more sustained because of an adverse distribution of power. This does not always mean a predatory distribution of power but could also be fragmented and subject to distributional conflicts. Attempting to scale in such a context without considering these structural drivers would be counterproductive. The coping capacities one needs to exit states of fragility might in fact be weakened. 

As a result, and paradoxically, one needs to ensure greater enforcement in a context where enforcement is significantly more difficult. In a number of situations, however, the solution to enforcement is top-down, using formal accountability structures and can even be securitised. This has usually led to limited success. 

While designing programmes for scaling it is important to understand these governance constraints and also assess them for feasibility. The scaling exercise has to have a response built-in as a response to address these constraints. Scaling has to be designed such that it builds resilience in a way weak governance mechanisms do not get exacerbated. 

But none of this is possible without a PE mapping or what we specifically do, that is mapping the power, capability, interests (PCI) of the relevant actors. Central to this analysis is the SOAS-ACE approach of horizontal enforcement where anti-corruption strategies include ‘insiders’ or those productive players interested in upholding rules in their sector for their own benefit. One of the most common reasons for corrupt behaviour is free riding where some actors break rules for a particular reason but others also violate rules given the rule breaking behaviour they observe around them. Free riding behaviour is likely to be even more widespread in fragile contexts making top down enforcement difficult. 

Looking for ‘insiders’ is more likely to work than top-down vertical enforcement that includes disciplining powerful, politically connected players who have little incentive to adhere to rules. Yet vertical enforcement is unfortunately the ‘go-to’ solution favoured in many fragile contexts. 

No enforcement problem can be considered in isolation. They have to be looked at as a collection of interactions and relationships that affect the behaviour (rule-following or rule-violation) of those directly being targeted by policy. A PCI analysis helps us understand this collective behaviour and assess the feasibility of solving this free rider problem, that is, to solve the collective action problem of rule violation. The question it asks is how can networked behaviour be influenced by policy in order to minimise rule violation. 

Power is relative and describes the likelihood of a specific actor winning against another in contests over resources. Capabilities describe how organizations make a living and organizations with productive capabilities are more likely to support the enforcement of productivity-enhancing rules. Interests depend on capabilities (productive or otherwise) but also on opportunities: productive organizations may have unproductive interests if they can get away with it.  

Contexts may be fragile but not anarchic. This methodology helps us identify why some resource flows are not captured and who the actors are that prevent this from happening and strengthen their capabilities through policy. In some cases such actors might not exist but it is possible to identify potential solutions involving some actors who have the power and capabilities to be interested in rule following and the next step would be to create these. In both cases what ensures rule enforcement is horizontal enforcement. Even where it seems vertical enforcement is in in action it is the peer to peer checking that allows for VE to be successful. 

By looking at PCI together, we arrive at an understanding of the organizations that are likely to engage in horizontal or peer to peer checking that supports the enforcement of productivity-enhancing rules. This mapping helps us understand who the rule followers are and not just who the rule violators are and this is the critical insight. And the next step is to follow through why this happens and strengthen it. This is even more critical in fragile contexts where we need quick policy implementation turnarounds. Horizontal enforcement helps us achieve resilience, sustainability, as well as the potential for scaling that may help exit fragility: as this case study from Nigeria will show. 

The Nigerian electricity grid was privatised in 2013 to improve transparency and governance but this exercise only led to politically connected and not necessarily technically competent bidders. As a result of being ‘too-big-to-fail’ the sector got continuous bail outs and finding which did little to change its performance as a result of the political connectedness of. In a country of 200 million with the largest number of people living under the poverty line and with high levels of unemployment and insecurity this was a significant constraint to broad-based sand inclusive growth.  

Due to the failure of grid-based supply, over 80 % of SMEs own or use a generator, or self-generate in Nigeria. They usually get no more than 4 to 5 hours of power a day and attempt to make do with a combination of costly self-generation and rule-violating behaviour like non-payment of bills, stealing from distribution lines, and using black market diesel for their generators. Attempts to enforce rules on them in such a context are politically unfeasible, and also untenable given their vulnerability in terms of losses of productivity. Costly self-generation is therefore both a cause and an effect of some types of corruption such as the theft of power from the grid and non-payment of bills.  

Alternative methods of improving power supplies for SMEs not only have to be economically viable, but they also have to address problems of corruption particularly at the level of non-payment of bills and power theft. In our research we established a willingness to pay by SMEs for power at prices higher than those notionally available in the grid, as they already self-generate at considerably higher prices (cost of corruption like paying off engineers, buying smuggled diesel (historically), cost of maintaining generator, cost of diesel). Many studies have established willingness-to-pay but this has not translated into implementation. In our case however, when provided with a solution that was modelled on supplying electricity only to their clusters the monitoring constraint disappeared as they were going to check others who were, on average, as powerful as them. Monitoring the performance of distribution companies was not plausible but monitoring each other provides a check on free riding behaviour.  

We also observed informal rule following arrangements among cluster members like payments for security or waste collection Our aim was to change how SMEs pay for electricity to a similar arrangement so that all benefit and therefore all pay.. This also ensures less variation in impact among SMEs, a key input for successful scaling. Traditionally measuring the impact of scaling has been by identifying the mean effect. But variations matter especially when populations are in vulnerable contexts. The more the variation in impact the less the chances that beneficiaries will adopt it. 

Horizontal checking based on the PCI mapping creates a self-sustaining virtuous cycle which enables sustainable rule-following behaviour, allowing for the developmental outcome we are hoping to achieve. Economic risks for potential investors in power generation are reduced because willingness to pay can be established together with horizontal checks that significantly raise the probability of payments. 

Scaling, especially for interventions related to livelihoods, also needs to be planned with financial sustainability (via de-risking) in sight because more often than not private finance is unavailable in these fragile and vulnerable contexts. Aid will run out; governments will run of budget. A mechanism that bakes in a self-sustaining financing model into the scaling exercise will help address governance sustainability, micro-level political sustainability and help mitigate conflict drivers. This de-risking afforded by the mapping is in itself a quasi-public good. In fact our approach is one that de-risks scaling but is equally relevant for scaling de-risking. Once this de-risking has been achieved private sector financiers are expected to make the required investment  

Our first step towards de-risking was a poof of concept which we wanted to follow up with modelling studies. But it transpired that SMEs needed to see this demonstrated in order to trust the model and so we are now moving ahead with a small-scale intervention in a furniture cluster in Abuja. The aim was to scale our outcome through a series of adaptive pathways given that more formal pathways to improve coping capacities were absent. This might seem like a lot of homework in fragile contexts where solutions are immediately needed, but not doing this homework locks in conditions for failure. 

To conclude, PCI mapping allows us to look for instances of horizontal enforcement that may be already working and needing to be strengthened or identifying potential strategies for incentivizing horizontal checking interests. Resilience is about building sustainability into governance mechanisms and this can only be the case when there is demand for rule following from actors interested in improving their productivity. In fragile contexts this provides the best chance at adherence and once this is established scaling is likely to be successfully. Otherwise the failures of vertical enforcement will persist.