Clientelism is often blamed for public service delivery failures in developing countries. While the top-down drivers of political support in exchange for service delivery for specific constituents is well-documented, local grassroots influence and the effect of democratic mobilisation by local communities are less well-understood. This column looks at the value of combining top-down influence with bottom-up community mobilisation to exert stronger pressures on improving anti-poverty programme outcomes.

Why in developing countries do some communities receive better access than others to government anti-poverty programmes that are ostensibly rule-based? For decades, the dominant framework for understanding this puzzle has been the literature on ‘clientelism’, which suggests that politicians manipulate public funds in order to cultivate political support. But are citizens simply passive recipients of funds distributed from above? The findings of my research suggest not. The communities that are best served by government are those that possess both top-down political connections to the ruling party, as well as high levels of grassroots local democratic mobilisation needed to place pressure on local politicians to deliver services.

Picking winners: A top-down story

To study this puzzle, I examined how the performance of India’s largest rural anti-poverty programme, the Mahatma Gandhi National Rural Employment Guarantee Act (MNREGA), varied across villages in Rajasthan – a poor, arid state in Western India where demand for the programme is particularly intense. Constituencies that were narrowly won or lost in the most recent state elections were compared through a survey of 2,250 households. In constituencies with close elections, results revealed that electing a ruling-party legislator significantly improved employment opportunities and wages received by local households under MNREGA – even though households in opposition-party constituencies were similar in terms of objective levels of demand for the programme. The gap was large enough to represent a roughly 0.83% increase in average household income, suggesting that political connections to the ruling party play an important role in shaping the livelihoods of ordinary citizens.

Looking at intermediate outcomes, the reason for this gap appears to be that villages in ruling-party constituencies received larger numbers of MNREGA projects – which are required to generate employment and wage payments under the programme. So far, this is consistent with the literature on clientelism, with ruling-party legislators utilising their political influence to deliver additional public projects to their home constituencies in order to cultivate political support.

Democratic mobilisation shapes service delivery

Looking within constituencies, however, the story becomes more complicated. It turns out that not all villages benefited equally from political connections to the ruling party. Instead, the benefits of electing a ruling-party legislator were overwhelmingly concentrated in “democratically mobilised” villages – places where the local village council was characterised by high levels of citizen participation and political competition.

To measure local democratic mobilisation, I averaged citizens’ own reports of levels of participation and competition in the local village council. Not only does there exist substantial variation across villages in the reported extent of local democratic mobilisation, but this variation plays a crucial role in shaping service delivery.

Why did ruling-party legislators target resources at democratically mobilised villages? It seems to be the case that democratically mobilised communities are better at placing grassroots pressure on local politicians and translating political connections into actual service delivery. This is also reflected in the content of village council meeting discussions. In democratically mobilised communities, a much larger share of households report discussions of substantive issues related to the performance of government programmes.

Combining top-down and bottom-up: A better outcome for development

The findings suggest that access to government services is a function of political connections, as the clientelism literature suggests, together with the internal democratic organisation of local communities, something that has not received enough attention. These findings suggest some novel combination of a top-down approach, as well as bottom-up institutional strategies are needed to strengthen service provision in developing countries in a manner that is consistent with democratic principles, and more likely to improve government programme performance

The implication of this research is certainly not that programmes should be shielded from political influence – that is a ´fool’s errand´. Instead, politicians should be provided with incentives to exert effort in the delivery of public services across constituencies on a broad basis. One counter-intuitive way to achieve this would be to institutionalise the role of locally elected legislators in the administrative process of allocating projects and benefits under large-scale anti-poverty programmes. This would a) provide opposition party legislators with greater influence, thus mitigating the exclusion of politically unaligned constituencies from access to programmes, and b) provide legislators with opportunities to claim credit for constituency service, thus linking electoral incentives to public service delivery.

Additionally, strengthening local democracy can encourage candidates from marginalised communities to contest local elections, encourage community attendance and participation at village meetings, and  preserve the rule of law and minimise violence and vote buying in local elections – all promising avenues through which to improve the performance of anti-poverty programmes on a “bottom-up” basis.

Twinning these two approaches could significantly strengthen the on-the-ground performance of anti-poverty programmes in developing countries.

This column first appeared on the IGC Blog.