The Centre for the Study of African Economies (CSAE) at Oxford University in collaboration with the Ghana Statistical Office (GSO) and the Tanzania National Bureau of Statistics (NBS) has been conducting a labour market panel survey of urban sectors in Ghana and Tanzania since 2004. There are now four waves of this survey covering the period 2004 to 2008. No survey was conducted in 2007 and data for that year depends on recall. Surveys were conducted in 2004-2006 and 2008. This dataset represents the 2008 survey. The survey collects information on incomes, education and labor market experience, household characteristics and various other modules for labor force participants (ages 15 to 60) in urban areas. For Ghana these areas span the four largest urban centers in the country: Accra (and neighboring Tema), Kumasi, Takoradi and Cape Coast. In Tanzania, the sample covers several of the largest urban areas including Arusha, Dar es Salaam, Iringa, Morogoro, Mwanza, and Tanga. The samples were based on a stratified random sample of urban households from the 2000 census in Ghana and the 2000 Household Budget Survey (HBS) in Tanzania. While the initial sample was household based, interviews were conducted on an individual basis, and the unit of analysis can be at the individual level. A total of 830 and 543 individuals were interviewed in the first round of the survey in Ghana and Tanzania respectively, which was conducted between October 2003 and June 2004. In Ghana only there was a follow-up survey of workers in Ghana’s manufacturing firms who had been surveyed from 1995. Thus the Ghana data contains those sampled on the basis of households and those drawn from firms. In using the data it is important to allow for the different basis of the two components of the sample. It is proposed to evaluate how access to micro-finance and processes of formalisation can impact on poverty by investigating two policies in Ghana and Tanzania. The first is the expansion of micro-credit services into randomly selected communities by several partner NGOs. The second is the implications of a process of formalising business structures currently underway in Tanzania. A three year panel of workers in the urban areas of both Ghana and Tanzania has already been created so a base line exists with which to compare the outcomes. The project is designed to establish how far these policies of credit and formalisation at the level of the individual can impact on poverty either directly by raising consumption or indirectly by increasing the productivity of employment opportunities available to the poor. It is proposed to re-survey the individuals in this panel in both countries to show how these changes have impacted on their livelihoods. Both interventions provide a unique opportunity to measure the causal impact of particular policies on poverty.

The survey instrument we used was a quantitative questionnaire focusing on individual incomes, educational background, health, fertility, social empowerment and other personal characteristics, as well as detailed household information on assets, housing, and consumption. For the three years 2004-2006 this survey had been administered to a random, cluster-based sample of approximately 1,000 urban households spanning eleven regions of Ghana and Tanzania. The survey had been conducted with the cooperation of the Ghana Statistical Service, the Tanzania National Bureau of Statistics and the Centre for Environmental Economics and Development Research in Dar es Salaam. Data collection had been carried out under the supervision of CSAE researchers in the field. These surveys are designed to allow comparison of the incomes of urban wage earners and selfemployed in both countries. From them we had already learned about factors determining these earnings. By extending this sample to incorporate beneficiaries of randomized interventions, we have created a unique opportunity to answer one of the questions set out above. First, with regard to identifying the beneficiaries of these flows (incidence), we can provide a profile of recipients and estimate the responsiveness of these financial flows to economic shocks. Second, by extending the panel to integrate an experimental design, we can measure the economic effects (impact) of microfinance relative to a baseline for each individual. Our proposal was to extend these surveys for an additional round. This was carried out in 2008 and with the help of supplementary funding in Ghana in 2009. As these samples are representative of the urban population, they provide an ideal context in which to investigate who gets micro-credits (incidence). In addition, the panel nature of the surveys allows us to rigorously evaluate the impacts of micro-credit by tracking randomly assigned 'treatment' and 'control' groups over time. Same interviews as those interviewed in 2006 panel survey.