DDI_MAR_2009_MCC-ES_v01_M Millennium Challenge Corporation 2015-05-26 NADA Version 1.0 (January 2015) Version 2.0 (April 2015). Edited version based on Version 01 (DDI-MCC-MAR-ME4-IPA-2013-v1) that was done by Millennium Challenge Corporation. Enterprise Support 2009-2013 MCC-ES 2009-13 Projet Soutien à l’entreprise MAR_2009_MCC-ES_v01_M Innovation for Poverty Action (IPA)

INDH

ANPME

OFPPT

Copyright 2012, Millennium Challenge Corporation. NADA Millennium Challenge Corporation APP Monitoring & Evaluation Division Independent Impact Evaluation Chaque série recueille les réponses des deux groupes traitement et contrôle indifféremment et pendant la même période. Edited clean data for internal use only Enterprise support Training Finance, Investment, Trade The evaluation will measure the specific contribution of the training and support program on key business outcomes. There have been no rigorous evaluations conducted on similar programs in Morocco. Thus, the evaluation will provide critical information to the APP and the Government of Morocco regarding the effectiveness of the Enterprise Support Project and ways in which it might be improved. The following are the principle research questions the evaluation is designed to address: Did the Enterprise Support Project contribute to an increase in revenues and profits of the participating entrepreneurs? - Did the program improve survival rates of businesses? - Did the program affect local employment? - Were the benefits generated by the project greater than project costs? The impact of the “AGR Support” program is measured by the differences between the AGRs in the control group and the AGRs in the treatment group for all the variables likely to be affected by the treatment. These differences are calculated by ordinary least squares regressions including stratum indicators as a control variable if the number of observations is sufficient, and - as often as possible - the value of the variable of interest at the time of the initial survey. The purpose of these control variables is to improve the precision of the estimators of the program's effect by reducing the statistical noise due to the intrinsic heterogeneity among the set of AGRs. Morocco The cities of Oujda and Sidi Ifni Individuals, enterprises 564 coordinators, as well as the 4 beneficiaries per AGR surveyed in the initial survey, for a total of 2,820 people. Sample survey data [ssd] Impact de l'appui aux entreprises et AGR sur leurs indicateurs financiers et socioéconomiques. Innovations for Poverty Action All the AGR coordinators were surveyed during the initial survey, but it was not possible to survey all the beneficiaries given the large number of beneficiaries per AGR. The evaluator sampled 4 beneficiaries per AGR: 1 beneficiary was randomly selected from a list of three beneficiaries who were eligible to receive training if selected in the treatment group, 1 board member was randomly selected from among the board members, and 2 beneficiaries were randomly selected from the list of all beneficiaries except for the 3 on the first list. The attrition analyses show that there is no significant difference between the initial sample and the final sample based on their baseline characteristics, i.e., the 550 coordinators in the final sample are representative of the initial 564 coordinators. Face-to-face [f2f] Moukawalati questionnaire For the purposes of fine-tuning the questionnaire, approximately 25 enterprises were visited in urban and rural areas. Cities and provinces included Kenitra, Tanger, Larache, Ksar Kebir, Temara, Khouribga, Fes, Safi, Casablanca, and Meknes. These enterprises represented a wide variety of sectors, thus as more was learned about the different contexts in which the programs are implemented, the questionnaires were improved to provide the best possible data. Overall, the prestests went very well and the effort put forth by the OFPPT and the ANPME to find willing entrepreneurs for the pretests was invaluable in refining the questionnaire. The questionnaire will collect the following information: -General characteristics of the entrepreneur: age, gender, level of education, field of specialization, professional experience, employment training, enterprise creation, income, wage (if applicable) and characteristics of the entrepreneur’s household (household size, assets, consumption, etc.) -General characteristics of the business: nature of the business, type of project (continuing/new), reason for creation, initial investment, sources of financing, type of clientele, number of potential clients, importance of competition, number of staff, internet access. -Revenues and costs: sales information, sources of income, costs, purchase of inputs, calculations of income flows and profits. -Competence: entrepreneurs’ knowledge and competence in the areas covered under the training such as management and accounting standards. -Entrepreneur’s ability: tests will be performed to evaluate entrepreneurs’ ability such as the digit span recall test, word problems and pattern recognition. Risk aversion and intertemporal choice will be measured through a field lottery experiment and questions regarding the interviewee’s preference for the present. AGR questionnaires As outlined above two separate questionnaires were created: a coordinator questionnaire and a beneficiary questionnaire. The same process of fine-tuning and pre-testing the questionnaires was carried out as in the case of the Moukawalati. For this purpose, we visited approximatively 15 AGRs in the following regions: El Jadida, El Kalaa, Guelmim, Tata, Errachidia and Al Hoceima. The coordinator questionnaire will measure the same business outcomes as noted above, but additional data will be collected concerning the structure and function of the AGR: -Organizational make-up: division of labor, decision-making process, shared means of production and commercialization. The beneficiary questionnaire will collect data on the participant’s specific role within the AGR: income, diffusion of the program on the members who do not receive the training and the wellbeing of his or her household. The evaluator resurved 10% of the respondents in order to ensure the reliability of the data. The backcheck for this first follow-up survey was conducted by telephone with 56 coordinators and 175 beneficiaries. The questionnaires used, which were shorter than those used in the field and had two different versions, went back over only the most important questions. The study is based on a formula of power calculation and therefore the weighting is done with the number of treated persons, the total number of eligibles, and the rate of take up fixed at 80%. The questionnaires completed by the teams were reviewed by the supervisors the same day as the survey in order to be able to report errors. They were then sent to the server at the Rabat office. Every day the server was synced to download the data. The updated database underwent a cleaning session performed by the IPA liaisons once a week. This second check enabled detection of inconsistent results, unjustified missing information (no filter), or surprising results. In the event that the errors could not be corrected directly, the backchecker was tasked with calling the survey respondent to have him/her clarify the problem items. All of the questionnaires were reviewed and, when necessary, the person surveyed was contacted again to check a specific item by means of a very brief call (less than 5 minutes). 98% of coordinators, 75% of recipients, and 72% of companies. Millennium Challenge Corporation Millennium Challenge Corporation http://data.mcc.gov/evaluations/index.php/catalog/157 Cost: None Innovations for Poverty Action. Impact Evaluation of the Income-Generating Activities (AGR) Support Pilot Project. November 2013. The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.