Impact of Behavioral Change Sessions on Attitudes Towards Work

Building on the Takaful and Karama cash transfer program (TKP), Forsa is a new pilot economic inclusion program implemented by the Ministry of Social Solidarity of Egypt. Forsa will provide short-term support to unemployed working age members of poor households by linking them to wage employment opportunities or providing training and assets to support self-employment. Before the actual implementation of the various interventions, eligible households are invited to a behavioral change session to motivate them to enroll in the program. Based on uptake data from the earliest governorate for implementation (which not included in the areas used for this analysis), approximately 67% of invited households sent a member to attend the behavioral change session.
Eligible households are either current beneficiaries of the TKP cash transfer program or rejected applicants just above the PMT threshold used for targeting the Forsa program who also fulfill the following eligibility criteria:
• Have at least one member of working age (19 to 55 years).
• Have a household head without formal employment and, if present, a spouse also without formal employment.
• Own less than 0.5 feddan (0.52 acres) of agricultural land.
• Not be a participant in any other transfer or asset program of the Ministry of Social Solidarity or other local or international non-governmental organization (NGO).
• Own no more than four medium-sized livestock or one large animal.
• Own a house with no more than one floor.
• Have no access to remittance income from a household member working outside of Egypt.
The objective of the behavioral change sessions is both to introduce more details about the Forsa program and to emphasize the benefits of financial independence and economic empowerment relative to financial reliance and aid dependency. The messages of the behavioral change sessions fall under four main themes:
Introducing the culture of work -. Participants were refreshed on the importance of work for improving their family’s standard of living and breaking out of the cycle of poverty.
Insisting that the continuity of Takaful transfers is not guaranteed – Participants were reminded that eligibility to receive Takaful transfers will end at some point due to limitations in the state’s budget. This insistence is consistent with the TKP program system of requiring recertification every three years and ending transfers for beneficiaries who are not deemed eligible during the recertification.
Emotional and religious encouragement. The trainers emphasized to the participants that Forsa is an opportunity for them to break intergenerational poverty cycles and prevent the transfer of their current deprivation and misery to their children. The trainers also incorporated a religious, focusing on how all religions praise hard work and the pursuit of general social and economic improvement.
Examples of success. The sessions also highlighted successful examples of people from their local communities who were able to break out of the poverty cycle by directing their efforts into income-generating opportunities.
Additionally, attendees are made aware that their Takaful cash transfers will not be stopped upon registration for Forsa – they will be sustained for at least 6 months to allow for a smooth transition into self-reliance. At the end of the session, households were required to make a decision about joining Forsa.
The behavioral change sessions are separated by gender and age and attended by up to 40 participants. They were supposed to include a standard presentation with video messages lasting four hours, however, it is uncertain how precisely these standards were followed during implementation.

2021-06-01

2022-03-15

• Income aspirations (wants)
• Income aspirations (needs)
• Risk and time preferences
• Locus of control, self-efficacy, grit
• Grit
• Preference between full time job in formal sector vs. Takaful transfer for 12 months, 9 months, or 6 months, 3months or 3 years of Takaful without Forsa
• Expectations about the continuity of Takaful transfers

The income aspirations outcomes are based on answers in the household survey about the income level needed for financial security and the highest income level the respondent aspires to. Risk and time preference outcomes each based on two piloted hypothetical questions about preferences between receiving different options, which will be used to categorize respondents into three groups: high risk aversion (patience), moderate risk aversion (patience) and low risk aversion (patience). Self-efficacy will be measured by degree of agreement with two selected statements from a standard module which worked well in piloting and are included in the baseline. Grit will be measured by the total score summing degree of agreement with a items from a standard module included in the baseline. The formal employment versus transfers preference is based on self-reported preferences for employment or transfers in five different scenarios with different durations for takaful transfers, with the outcome being defined as the switching point between preferences for work and receiving transfers. The expectation about continuity of transfers is the answer to a direct question in the baseline.

This study is embedded within the baseline data collection of a cluster randomized controlled trial to measure the impact of Forsa.
For this study, we randomized at the household level which households received an invitation to the behavioral change session prior to the baseline survey data collection in January-February 2022. The exact timing of the behavioral change sessions in each area depended on the pace of implementation and varied from June to December 2021. Households randomized into the control group for the behavioral change session will receive an invitation in March 2022.

Randomization of invitation timing was done at the household level among the list of all eligible households intended to be targeted.
The process of randomization differed slightly by governorate as it was not initially clear when the baseline data collection could take place relative to the behavioral change sessions. In four of the seven governorates, Luxor, Assuit, Beni-Suef and Menia, randomization was applied in two steps: in the first step, half of the eligible households were randomized into the control group. The other half (treatment list) was sent to the implementing organization to be targeted in the first two weeks of implementation. After the first two weeks of implementation, more households were needed for treatment, so another randomization round was applied (the second step) and only 50 eligible households remained as the control group. In the remaining governorates, randomization was applied in one step by randomly selecting 50 eligible households for the control group. As a result of the two randomization processes, there were fewer potential replacement households in some governorates, and we anticipate this may lead to some loss in sample size as not all households listed in the administrative data will be able to be located. The fewer potential replacement households in some governorates are identified to be in 18 villages in Luxor, Beni Suef and Menia.

In office by computer (reproducible Stata code)

Household

No

3312 households

3312 (24 households per community in 138 communities) Takaful beneficiaries: 2,208 (16 households per community in 138 communities) Rejected Takaful applications: 1,104 (8 households per community in 138 communities)

1506 control, 1506 treatment

MDE is 0.14 standard deviations for the given sample size and number of sites assuming perfect compliance and 0.1 effect size variability. (We do not have any confident basis for anticipating the mean and variance of our outcome variables, so prefer to specify in standard deviations.) Expected take-up of 0.67 implies needed sample size will be 1/(0.67^2) larger for same power, so accounting for this, MDE is 0.2 standard deviations. Imperfect compliance may further increase this MDE.