Faculty Voices: Rebuilding the Social Compact in Pakistan

This is part of a series in which we highlight recipients of SAI’s Faculty Grant program.

Rebuilding the Social Compact: Urban Service Delivery and Property Taxes in Pakistan
Principal Investigator: Asim Khwaja (Harvard Kennedy School)
Co-Investigators: Ben Olken (MIT), Adnan Khan (LSE/IGC) 

This document summarizes pilot work conducted on the Social Compact project from July 2015 to June 2016, supported in part by the Harvard South Asia Institute. This includes the following activities:

1. Defining the menu of urban services
2. Pilot exercise
3. Government meetings and approvals
4. Field visits
5. Updates to intervention design
6. Neighborhood construction

Background and relevance

The social compact between citizen and state – whereby a citizen pays taxes and receives (public) goods and services – is a critical link in the development process. This link is especially salient in the context of local governments and a significant metric on which they are judged. However, if citizens perceive little benefit from their tax payments, or local services are disconnected from local decision-making, this link can be broken. This can create a vicious cycle where citizens do not receive high quality services because resources are limited by low levels of local tax revenue, and the low quality of services leads to a low willingness to pay taxes, as well as a broader lack of trust in the state.

This study seeks to examine how to break this cycle through a series of reforms that strengthen the link between the provision of local services and local property tax collection in urban Pakistan. We aim to increase the link between taxes paid and local services received by enhancing the linkages between a) local preferences and the types of local urban services provided, and b) the amount of local tax revenue and amount of local services provided. The goal is to reduce evasion of local property taxes, increase the quality of local services, make local urban services more responsive to local needs – and ultimately to help repair the social compact between citizenry and the state.

Pilot work

1. Defining the menu of urban services

To identify which services are most needed in the target areas, we collaborated with the Punjab Municipal Development Fund Company (PMDFC) and the local Town and Municipal Administration (TMA) offices to determine a shortlist of needed services based on need and feasibility. Through field visits to potential target areas, we also elicited preferences from citizens and narrowed down the list to the following four services that were both feasible and likely to respond to citizen needs: clean water access, sanitation, street repair, and street lighting.

Since the precise service to be provided depends on amount of revenue available (which is not known ex-ante), we identified a range of options within each category in collaboration with PMDFC. These options are as follows:

• Clean Water: Replacement of pipes, installation and maintenance of a local water filtration system
• Sanitation: Improved trash collection, street cleaning, installation of trash dumpsters, and construction of gully grating chambers (which separate silt from sewage)
• Street Repair: Pothole repair on neighborhood streets, street resurfacing/soling, replacing manhole covers
• Street Lighting: installation of one more or lamp posts, maintenance of lamp post and lighting

We also collaborated with PMDFC and the TMA in Toba Tek Singh (of which one of our pilot locations, Gojra, is a member) to determine the cost of implementation for each service. The Toba Tek Singh TMA also provided official budget documentation with a detailed breakdown of service costs in Gojra, which allowed a precise appraisal of service costs.

2. Pilot exercise

To refine implementation challenges before the main project rollout in July 2016, we launched a pilot in the fall of 2015 in 20 urban neighborhoods, with 100-200 properties each, in two cities: Lahore and Gojra. These municipalities were selected after extensive consultations with the PMDFC, various TMAs, and other government departments based on characteristics such as maintenance of detailed budgeting and expense records and monitoring of local services through a performance tracking system. Conducting the pilot in Lahore and Gojra allowed for testing the reforms in two very different settings: one large city, Lahore, where service provision is outsourced to other public and private companies and one small city, Gojra, where service provision occurs almost entirely by the local government (TMA) itself. The pilot therefore captures differences in responses to the reforms due to the municipality setting, as well as highlighting the logistical issues that may vary across municipalities due to differences in how services are provided.

3. Government meetings and approvals

Key government stakeholders at all levels approved the proposed interventions. This includes the Chief Minister Punjab, who has formally approved this project and set up a Steering Committee to oversee the project implementation. We have also met and are actively collaborating with the Local Government Department, Excise and Taxation Department, several Town and Municipal Authorities, the Punjab Municipal Development Fund Company, and Urban Unit (responsible for GIS mapping of properties and digitizing the tax billing process). These departments are willing and keen partners for the project, and the study team has met with their representatives and senior officials. Our policy partners have helped shape project design through our meetings and have shared relevant administrative data to support project implementation. These efforts have paved the way for collaboration during the main project rollout in July 2016.

4. Field visits

Prior to implementing the pilot exercise, we conducted field visits to urban areas across Punjab to collect detailed qualitative information that was used to develop the project design. Each field visit included lengthy, open-ended discussions with residents and shopkeepers, as well as qualitative assessments of the quality of local services. Specifically, respondents were asked about their opinions towards the government, their interactions with the government, their assessment of the need for particular services, their receptiveness to a scheme tying citizen preferences to service delivery, and more. Forms to collect data on citizens’ service preferences were also piloted and refined to determine which format could collect high quality data with ease and accuracy.

5. Updates to intervention design

In the past year, we updated the list of interventions, removing the matching grant component and adding a pure local allocation component. The updated list is shown below:

1. Information and Preference Elicitation. Tax staff will inform citizens of the tax-service linkage and give them a more direct voice in how their taxes will be utilized by showing them a short video about how local taxes are used. Tax staff will then solicit citizens’ preferences on which type of local goods should be prioritized in their neighborhood. The results of this preference elicitation will be shared with the local government in an effort to improve the allocation of services.
2. Local Allocation. In the status quo, revenue is collected from administrative tax units and transferred to local governments that allocate these to city-level services. To strengthen the link between taxes paid and services provided, this intervention will require local governments to allocate a portion of property tax collected from a neighborhood (tentatively 35%) to that same neighborhood.
3. Preference-Based Local Allocation. This intervention combines the previous two. By both eliciting citizen preferences (#1) and requiring local governments to allocate funds to the neighborhood (#2) in accordance with these preferences, it seeks to make the tax-services link even more salient and credible. Citizens will be informed of this earmarking via a different video in the app, and the subsequent service expenditures will be carried out in their locality.

The updated sample allocation is as follows:

In addition to these main treatments, we will implement a variant of the preference-based local allocation intervention where local leaders will mobilize the community to help enhance the tax-service link. Finally, we will randomize the content of the information and preference elicitation at the property level to help understand how to make the information credible and how to best get citizen voice. In particular, we will vary who will deliver the information in the video (a high ranking politician or bureaucrat), whether taxpayers rate the quality of existing services in addition to providing preferences on new services, and whether taxpayers are given the opportunity to provide unstructured feedback to the government.

6. Neighborhood construction

The pilot has allowed us to develop our neighborhood mapping and construction process. We collected detailed geo-located administrative data on property-level tax collection from the Excise and Taxation Office and the Urban Unit in the pilot areas, which we used to define comparably sized neighborhoods and which will allow us to track outcomes at the property level. We used data on the cost of provision of various services to determine the neighborhood size such that revenue collected above a historical benchmark from a neighborhood would be able to finance service provision in the preference-based allocation intervention areas; this information informed our study design. The neighborhood construction process is currently under way, and is scheduled for completion prior to the full-scale rollout of the intervention.