The Role of Information, Accountability, and Resource Gaps in Explaining Poor Urban Services Quality in Addis Ababa and Its Rapidly Urbanizing Surroundings

The goal of our proposed project is to explore the role of three potential, non-mutually exclusive, reasons for why the responsiveness of local officials to local needs is low: (1) they lack information on the citizen’s priorities; (2) they lack the incentives to respond to citizen’s priorities because they are not held accountable for citizen’s well-being; (3) they lack the autonomy (i.e. the power to access the resources) to respond to citizen’s priorities.

We propose to explore these issues by (1) surveying local officials (both elected officials and bureaucrats) to gauge the extent to which there is a mismatch between what they perceive as the local community’s priorities and the actual priorities, as well as between their stated priorities and what they think are the community’s priorities; and (2) delivering a “report card” treatment intervention in which we will report summary information on citizen’s concerns and needs. The report cards were shared with officials at different levels of the administrative hierarchy. We will then trace the extent to which the report card intervention affects budget allocations and policy choices, and how this depends on who is targeted by the intervention.

In order to further understand the mechanism through which this type of information sharing can influence policy choices and bureaucrat action, we also delivered "placebo" report cards to a subset of administrative units in Addis Ababa. Rather than display specific statistics, like the standard report card, these placebo cards simply described the project and the team's intention to make return visits over the next few years. This will allow the team to understand whether observed impacts are driven by the promise of future monitoring or the content of the report cards.

Detailed description of the intervention:

The administrative unit of interest is the woreda. In Oromia, woredas are governed by zones or directly by the region. Some large towns constitute their own woreda. In Addis Ababa, woredas are overseen by sub-cities. Each woreda has an top administrator and 15+ sector bureaus, each of which has a manager.

- 5 report cards were produced for each woreda, covering the following 5 domains: Education; Health; Small and Medium Enterprise Development (SME); Water; Electricity; and Jobs. Report cards display data on the public's satisfaction, usage, and preferences related to the domain. This data was collected from a representative sample of individuals and firms in 2016 as part of a larger data collection initiative. Each report card was adapted to ensure statistics provided were based on a minimum number of observations (households and firms). Consequently, some statistics were calculated at the zonal or sub-city level rather than woreda level. When zones in Oromia contained both urban and rural woredas, and statistics were calculated at the zonal level, the statistics for rural and urban woredas in the zone were calculated separately. Woredas in these zones would then receive the statistics that corresponded to their own category (rural or urban). The report cards were printed on a page with two sides. One side provided general information about the data collection effort done by the research team. The other side provided the statistics.
- Placebo report cards were produced. Those only had the general information about the data collection effort and no statistics.
- In woredas sampled for the report card (RC) treatment, an initial set of two RCs were delivered to the sector bureau manager in two (randomly selected) of three sector bureaus (Education, Health, Small and Medium Enterprise Development (SME)). A few months later, the top administrator received the two woreda-level report cards on jobs, water, and electricity, as well as the sector-level report card for the sector bureau that did not receive an RC during the initial report card delivery period. For example, if initially the education and health RCs were delivered to sector bureaus managers, the administrator received the SME card. Hence, all treated woredas received the same set of 5 report cards overall, but the level (sector manager vs. top woreda administrator) at which a card was introduced varies randomly across RCs for the initial set of RCs.
- In woredas sampled for the Placebo treatment, placebo cards were similarly delivered to the manager of two of the three previously mentioned sector bureaus, as well as the top woreda administrator.

2017-09-25

2018-04-04

Officials' awareness of citizens' concerns, policy choices, budget allocations, administrator engagement

Official awareness will be measured by comparing citizen priorities with local administrative office priorities, and comparing citizen priorities predicted by the official with actual priorities reported by citizens in surveys. Policy choices and budget allocations will be taken from official administrative records. Administrator engagement will be measured based on response to the feedback forms.

Public satisfaction and priorities, community engagement

Community engagement will be measured by measuring administrative outreach through public meetings, announcements, and changes in responsiveness to citizen complaints. Public satisfaction will be measured through repeated questions during the second wave of surveys with individuals in 2019.

Stratifying by zone/sub-city and woreda budget per capita, woredas were assigned to treatment and control groups (29 control, 29 treated) split between Addis Ababa and the Oromia Region. Treated woredas were then assigned to receive report cards in 2 sectors randomly selected from a list of 3. These report cards will be delivered to the manager in charge of a selected sector bureau.

For the placebo report cards, 89 woredas in Addis Ababa for which no household or firm survey data is available were stratified by sub-city and total woreda budget. These were then similarly split into treatment and control groups (45 control, 44 treated). Placebo report cards will be delivered to sector bureau managers in the selected bureaus as well as to the woreda administrator.

A second survey was conducted with all woreda administrators a few months after the report cards were delivered to sector bureau managers. Following this second survey, administrators in treated woredas were given a set of broad report cards for service areas that are managed by multiple sector bureaus. The administrator also received the sector-level report card that was not delivered during the first delivery period (where only 2 of 3 bureaus managers received report cards). Finally, all bureau managers received a feedback form that could be sent to the project with their reaction to the survey. The feedback forms in treated woredas included additional questions about the utility and relevance of the delivered report cards.

An endline survey will be conducted with all woredas administrators and sector bureau managers approximately 1.5 years after the initial report cards were delivered. At endline a host of administrative data will be collected (in particular, budgets).

Done in office by a computer

Woreda, woreda-sector

No

147 woredas
882 (147*6) woreda-sectors

Approximately 1000 administrative officials

Standards report card treatment: 29 woredas
Placebo report card treatment: 44 woredas
Control: 74 woredas