Periodic Reporting for period 1 - QUELIMANE (Integrating Rural Migrants in Cities - A Field Experiment in Mozambique)
Reporting period: 2021-06-01 to 2023-05-31
The problem being addressed in this project is that of peacefully integrating rural migrants into urban polities. This is important for society because rural-to-urban migration is one of the most well-proven tools researchers are aware of for fighting poverty. Sub-Saharan Africa is the world’s fastest-urbanizing region. The movement of people from the countryside to denser cities will likely create progress toward some of the Sustainable Development Goals in the medium term, especially SDGs 11 (Sustainable Cities and Communities), 13 (Climate Action), 8 (Decent Work and Economic Growth) and 5 (Gender Equality). However, rural-urban migration can also cause disruption and conflict in the short term. Furthermore, many autocratic national governments oppose urbanization because it increases the cost and difficulty of controlling citizens through patronage and intimidation. Therefore, the responsibility often falls on city-level governments in the developing world to peacefully integrate new migrants with existing residents, and there is an urgent need for evidence on how to do so. This project will help meet this need in the context of the Mozambican city of Quelimane (pop. 350,000). Quelimane has created a program for rural migrant integration that includes job matching, training and education, financial services, and tax incentives.
The overall objectives of this project are to conduct and report on a randomized controlled trial of a local government program intended to integrate rural migrants into a city in Mozambique.
The overall objectives of this project are to conduct and report on a randomized controlled trial of a local government program intended to integrate rural migrants into a city in Mozambique.
In late 2021, a team of survey enumerators conducted a baseline survey of over 2000 migrants, over 1000 existing residents, and around 500 chiefs. This survey asked about income, employment, family structure, education, migration plans, attitudes toward migrants, politics, and more. The roughly 500 blocks in the city were divided up randomly into three groups. The first treatment arm (“T1”) would receive the proposed migrant integration package, administered technocratically by program officials on behalf of the municipality. A second treatment arm (“T2”) would receive the program, but administered alongside the block chief and with the block chief’s involvement. The control group would receive no program.
After a few months, a follow-up sampling procedure was conducted to increase the sample size, and program administrators began conducting visits among the migrants sampled in T1 and T2 blocks. These visits consisted of information about municipal services such as health and education; help setting up mobile money accounts and learning to send remittances via mobile money; and a job matching program. The job matching program was created by our research team, and involved our enumerators conducting a job census of the city, knocking on the door of every business establishment that could be found to ask if they were hiring and whether they would consider migrants for the job. Phone numbers from jobs that fit the desired jobs of treated migrants were then provided to the migrants; this constituted the job matching program. This process was repeated various times, with new job censuses and new rounds of visits by program administrators.
In mid-2023, a “midline” survey was conducted among all migrants surveyed at baseline.
The next steps are to analyze data from the midline survey, conduct a final (fifth) round of treatment visits, and conduct a final endline survey, in November 2023 just after municipal elections.
As of the writing of this summary, because the data collection period has not concluded, results of the intervention are not available. Therefore there are no results to exploit or disseminate. Within the next few months, the data collection will complete, and the results will be revealed. After that, they will be disseminated.
That said, because the population we study is so rarely surveyed, even the descriptive statistics from the baseline survey are instructive. In this paragraph we provide some of these descriptive statistics. Among the 2320 migrants we surveyed in the city – consisting of people who had moved to the city in the preceding 18 months – 22% had some paid occupation, highlighting the need for labor market integration. 25% had sent money home in the previous 12 months, highlighting the important role of remittances in this population and their continuing links to their place of origin. 35% said they had asked a local leader for help with some issue since moving, highlighting the capacity for the local government to alleviate labor market frictions.
So far this project has not led to dissemination and exploitation activities, as there are no final results for dissemination yet.
Communication activities: 1 workshop organized, 1 website.
After a few months, a follow-up sampling procedure was conducted to increase the sample size, and program administrators began conducting visits among the migrants sampled in T1 and T2 blocks. These visits consisted of information about municipal services such as health and education; help setting up mobile money accounts and learning to send remittances via mobile money; and a job matching program. The job matching program was created by our research team, and involved our enumerators conducting a job census of the city, knocking on the door of every business establishment that could be found to ask if they were hiring and whether they would consider migrants for the job. Phone numbers from jobs that fit the desired jobs of treated migrants were then provided to the migrants; this constituted the job matching program. This process was repeated various times, with new job censuses and new rounds of visits by program administrators.
In mid-2023, a “midline” survey was conducted among all migrants surveyed at baseline.
The next steps are to analyze data from the midline survey, conduct a final (fifth) round of treatment visits, and conduct a final endline survey, in November 2023 just after municipal elections.
As of the writing of this summary, because the data collection period has not concluded, results of the intervention are not available. Therefore there are no results to exploit or disseminate. Within the next few months, the data collection will complete, and the results will be revealed. After that, they will be disseminated.
That said, because the population we study is so rarely surveyed, even the descriptive statistics from the baseline survey are instructive. In this paragraph we provide some of these descriptive statistics. Among the 2320 migrants we surveyed in the city – consisting of people who had moved to the city in the preceding 18 months – 22% had some paid occupation, highlighting the need for labor market integration. 25% had sent money home in the previous 12 months, highlighting the important role of remittances in this population and their continuing links to their place of origin. 35% said they had asked a local leader for help with some issue since moving, highlighting the capacity for the local government to alleviate labor market frictions.
So far this project has not led to dissemination and exploitation activities, as there are no final results for dissemination yet.
Communication activities: 1 workshop organized, 1 website.
Internal migrants are an inherently difficult population to study because of their high mobility. We have therefore utilized new survey techniques to survey respondents remotely via phone interviews as well as in person, taking care to incorporate the latest best practices which were developed during the COVID-19 pandemic. Another dimension in which our project pushes forward the state of the art is in our behavioral measures of political economy variables such as corruption, civic cohesion, and trust. We aim to study not only how the program affects recipients, but also how it affects the local chiefs tasked with administering the program. Therefore we have implemented a series of behavioral "games" and other activities with real stakes in order to test the likelihood that local chiefs give valuable jobs to relatives rather than qualified individuals; how well they know the migrants in their block; and how much they are willing to make costly sacrifices in order to harm (or help) migrants and non-migrants on their block.
We are now in the process of planning the final round of data collection, so it is premature to speculate about what the results of the intervention might be. However, we can say that our study will provide one of the deepest and most comprehensive snapshots to date of rural-to-urban migrants in Mozambique, or indeed anywhere. Our results will provide solid evidence on whether programs of the type we study can succeed in integrating rural migrants into urban labor markets. Equally important is the light the project will shed on the political constraints and/or opportunities involved with serving internal migrant populations. A conventional wisdom holds that catering to migrants' needs can only be done at the expense of the goodwill of the existing population; we aim to test whether it is possible to in fact make it politically incentive compatible through the kind of program we evaluate. Insofar as migration represents one of the most powerful anti-poverty programs currently known in the literature, this could have significant socio-economic and societal implications by illuminating the political opportunities and tradeoffs of appealing to migrants for local leaders around the world.
We are now in the process of planning the final round of data collection, so it is premature to speculate about what the results of the intervention might be. However, we can say that our study will provide one of the deepest and most comprehensive snapshots to date of rural-to-urban migrants in Mozambique, or indeed anywhere. Our results will provide solid evidence on whether programs of the type we study can succeed in integrating rural migrants into urban labor markets. Equally important is the light the project will shed on the political constraints and/or opportunities involved with serving internal migrant populations. A conventional wisdom holds that catering to migrants' needs can only be done at the expense of the goodwill of the existing population; we aim to test whether it is possible to in fact make it politically incentive compatible through the kind of program we evaluate. Insofar as migration represents one of the most powerful anti-poverty programs currently known in the literature, this could have significant socio-economic and societal implications by illuminating the political opportunities and tradeoffs of appealing to migrants for local leaders around the world.