Partnership Lessons 5: Malawi Taxes for Public Services with the Institute for Public Opinion and Research

EGAP Researcher: Jake Bowers, Matthew Winters

Other authors: Nuole Chen, Christopher Grady, Boniface Dulani, Mwayi Masumba

Partners: Institute for Public Opinion and Research

Geographical region: Africa

Preparer: Melanie Phillips

June 15, 2020

Background

This study was conducted in Zomba, Malawi, in the context of continually expanding city boundaries that have resulted in citizens being newly incorporated and expected to pay property taxes (e.g. taxes for funding public services). There is reluctance and refusal to pay property taxes which is often motivated by inadequacies in service delivery that, in turn, creates further failures in service provision—a cycle that can debilitate city governance’s growth and institutionalization. This project focused on solid waste service collection, tax collection, and citizen attitudes. In the study’s sample, there is variation in experience with waste collection though none of the wards received waste collection as a public service. The intervention’s focus was, in the context of expanded waste collection, to encourage citizens to buy into public services and further institutionalize delivery by increasing continuous payments of property taxes. The aim of this study then, was that by demonstrating the worth of a public good through the provision of either information or the good itself could change local attitudes that form barriers to tax collection.

The field experiment was designed to assess the effect of incentives and information about paying property taxes and receiving solid waste collection on a household’s likelihood to pay taxes and continue paying taxes in the future. Additionally, it was designed to look at how paying property taxes and receiving waste collection affected a household’s relationship with the local city council. Lastly, the study looked at how information spread in communities who did and did not receive the informational treatment (i.e. spillover of treatment effects). The project focused on three key components: inadequacies in information, barriers to payment, and lack of service provision in Zomba. To increase information and decrease barriers to payment, the researchers created a brochure with information about services funded by property taxes and payment options for property taxes that were distributed to randomly-selected households in Zomba. Lastly, to increase service provision and thereby an individual’s experience with service provision, the project provided a skip loader truck and eight skip bins.  The skip bins were placed around the four wards and residents dumped their waste in the skip bins, which were then picked up by the skip loader truck and hauled off to the landfill.

The team began dissemination events sharing the results of the project in March 2020. These events included communities involved and affected by the intervention, as well as the Zomba City Council. However, while six dissemination meetings were originally scheduled, only three were completed as the research team’s efforts were interrupted by COVID-19 and will have to be rescheduled for later in the year.

In the sections that follow, this brief summarizes responses from an interview with the implementing partners on this project, Mwayi Masumbu (Associate Researcher) and Dr. Boniface Dulani (Senior Partner and co-author). The interview discussed the decisions to partner with the Institute for Public Opinion and Research (IPOR), how this collaboration worked during the implementation of the project, and their recommendations for future research.

Decisions to Partner with the Research Team

The project was implemented by IPOR in Malawi. IPOR is a well-established research firm in the country that has been involved in numerous academic projects in multiple sectors of academic and public policy research. The research team, made up of four academics at the University of Illinois at Urbana-Champaign, interviewed a small handful of firms when visiting the country, and ultimately selected IPOR based on their experience and fit with the project. Further strengthening the relationship between IPOR and the research team was the fact that Dulani, the senior partner at IPOR, was brought onto the project as a co-author as well as the director of implementation. This overlap in the role of researcher and implementor not only improved the research design and the understanding of the findings, but also the success of the intervention.

The design of the intervention for this project is unique in that it provides a public good to its participants: waste collection. It was this facet, in particular, that drew the interest of IPOR. Dulani noted that a common question most participants asked when being invited to participate in a study is “How does this benefit me?”. The answer to this commonly being that, while this might not directly benefit you right now, the knowledge from the project could benefit many in your community. This is a common dilemma facing many surveys and experimental projects, but this intervention could actually point to a benefit that participants would be provided with immediately. The service provision component, Dulani and Masumbu noted, excited them to work on this project but also improved the reputation of IPOR and survey research as a whole among participants, as well as improved their relationship with local city council members.

Furthermore, Dulani pointed to the publication opportunities that will emerge from this project as another benefit of the collaboration between the researchers and IPOR.  A working paper about the overall results of the experiment is already available and there are papers planned to study the relationships between using city services and changes in attitude, and to study a public goods behavioral game implemented during the project, all with Dulani as co-author.  Additionally, this project will produce multiple briefs, like this one, and reports for IPOR, which will both help with the dissemination of results but also bring publicity and capacity building for IPOR.

It is important to note, however, that while Dulani and Masumbu highlighted the many benefits of this collaboration for IPOR as an institution, it was not without costs. Dulani explained that this project was unique in that the budget did not account for the normal overhead costs that IPOR charges to manage projects. These fees go to funding the center itself and its researchers, all of whom were utilized to the benefit of the project while in Malawi. While the decision to forfeit overhead costs was made with the many benefits of this project in mind, it is important to note that if this was to be a norm for research projects, institutions like IPOR would not grow and likely would cease to exist.

Working with the Team

The collaboration between IPOR and the team went smoothly, as noted by Masumbu, with a clear division of responsibilities. During the entirety of implementation, research team member Chris Grady  was on-site in Zomba at the IPOR offices.  Research team members Matt Winters and Lula Chen were briefly in Zomba at the start of the project for initial conversations, and Lula Chen was briefly in Zomba at the end of implementation to collect administrative data at Zomba City Council offices. While Masumbu managed the enumerators in the field who were responsible for delivering the treatments as well as the pre and post-treatment surveys, Grady managed the back end ensuring data collection was running smoothly. IPOR was instrumental in providing experienced enumerators, gaining necessary permissions from local authorities, and providing necessary access to local communities through their language abilities and local legitimacy.

The implementation ran into expected roadblocks that required flexibility for which both IPOR and the research team adjusted for successfully. For example, while the research team originally planned to provide tablets as a cost-saving measure, Dulani noted that it became clear that the tablets provided would not be able to collect the necessary information, specifically GPS data, needed to conduct the project successfully. Quick to adjust, IPOR provided their tablets and the project continued unphased by this misstep because of the strength in collaboration. Further, the research team first assumed that providing a waste collection skip loader truck and skip bins was an easy task, but this task required much assistance from IPOR due to local city and country laws. Ultimately, IPOR acquired a truck in South African and drove it to Malawi, a task which IPOR accomplished while caught in the middle of Cyclone Idai.  This drive caused some damage to the truck and it had to be fixed upon arrival in Malawi, which IPOR arranged.  Any researcher who has implemented a project near the size of this has come to expect such hurdles, but the continued success of this project was the result of flexible collaboration and a well-established local partner.

While the relationship between IPOR and the researchers went smoothly, there is room for future improvements through capacity building. Specifically, while IPOR and its team of well-experienced researchers managed the implementation of the project, most were left in the dark in the ways of design and analysis. Masumbu was clear to point out that Grady made efforts to teach the enumerators some survey programming skills through an informal workshop using SurveyCTO and CommCare.  Dulani noted that design and analysis training could be something formally incorporated into future project budgets to increase capacity building.

Dissemination of Findings

Dissemination meetings for the findings were held in March 2020. Until interrupted by COVID-19, they held three separate meetings with local community members. Participants of the experiment were invited, but word of mouth led to a large attendance from the community at large. Notably, Grady commented on the fact that attendees lined up around the block, and questions and discussion went long past the originally allocated time. The disseminations themselves focused on descriptive statistics from the project. Specifically, the number of people who had access to waste collection, the rate of discussion about waste collection at community meetings, the services that participants reported interest in receiving, and whether the provision of waste collection affected participants’ willingness to pay property taxes and their overall attitudes towards the city government. Extending past the initial presentation of the results, Grady highlighted that the dissemination meetings provided a space for community members to engage with IPOR and the researchers about the research design, the community sample chosen, and the choice of questions included in the survey instrument.

Three more dissemination events are planned to be conducted in the future. These events will target political actors and academic actors—Zomba City Council, the Chancellor of the University of Malawi, and Lilongwe City Council. While city council members were originally invited to and attended the first dissemination event, Masumbu pointed out that it quickly became clear that when working on a contentious topic, like property taxes and service provision, having city officials  in attendance can derail conversations away from the project. While a discussion between community members and city council members could prove fruitful, it was not the ambition of these dissemination events to put the city council in the position to directly respond to the findings and community complaints. From this experience, Masumbu highlighted the importance of strategically targeting certain groups, such as the community or the city council, and holding multiple events in order to successfully engage the audience with the results of the project.

Recommendations

It is clear from the discussion with Dulani and Masumbu that the collaboration between the research team and IPOR was ultimately successful. The project greatly benefitted from IPOR’s experience and capacity in Zomba, and IPOR’s reputation with the local city council and community improved through this service provision study. Dulani noted that they look forward to the future publications that will emerge from this project and the rescheduled dissemination events. IPOR proved a vital resource beyond mere implementing partner, providing local access and legitimacy, in addition to acting as a safety net when expected problems arose. However, for institutions like IPOR to continue to grow and be the much-needed institutions that academics in this line of research rely on there are two recommendations that come from this partner. First, as Dulani described, is that while the cost of survey research is high, projects of this nature should account for overhead costs of their implementing partners in budgets to ensure the survival of these research institutions. And second, as Dulani pointed out, projects like this should incorporate more formal trainings on research methodologies to encourage capacity building of the implementing partners.