Measuring support to local governance in Pakistan

GIZ helped to develop and pilot a participatory development planning approach for the new local governments (the intervention). The main objectives of the intervention were to strengthen the involvement of citizens in local development planning and to increase the satisfaction of citizens with information on development planning received from government actors. A quasi-experimental research design with an innovative sampling method was integrated into the project. Three rounds of survey data from intervention and control areas were collected to measure the effectiveness of the participatory development planning approach. The analysis shows that the intervention was highly effective in raising citizens’ levels of satisfaction with information on development planning, but it had no clear positive effect on people’s knowledge of development planning processes. While the statistical evidence of the effect of the intervention on citizens’ involvement in development planning is mixed, the acceptance and scale-up of GIZ’s approach by the government and development partners speaks for its success. The provincial government and partners appreciated the practicality of the intervention as well as its appropriateness in the context, together with the data on citizens’ satisfaction. Therefore, in 2017 Khyber Pakhtunkhwa adopted the approach as the province’s official development planning methodology in 12 out of the 25 districts that implemented an EU-funded programme. The EU, UNDP and SDC supported the scale-up of the participatory planning procedure to hundreds of local councils, including in Punjab province. In 2017, a technical assistance-based successor programme by GIZ adapted and expanded the original research design to cover additional thematic and geographic areas. A review of the quasi-experimental research yields several lessons for future analyses and programming, including on reporting the robustness of results, communication of research findings and strategic choices about research methods that are informed by programmatic learning interests. One of the main conclusions is that while the research approach was state-of-the-art and scientifically rigorous, it would have benefited from a more transparent and proactive communication of findings for programme managers and partners to make full use of the results.