Impact Evaluation of Interventions to Expand Social Security in Jordan

Disseminating information about social security through different modes of communication.

Context

Estidama++: International Labour Organisation (ILO), in collaboration with Social Security Corporation (SSC), has launched a programme called Estidama++ in Jordan. Estidama++ aims to extend coverage to vulnerable workers by providing a contribution subsidy to employers and cash support to workers. Employers can apply to receive a subsidy of 30 JOD (42.24 USD) per month to their social security contributions for their employees who they registered during the programme time window. Newly registered employees also receive cash transfers valued at 100 JOD (140.75 USD) every three months. The subsidy and cash transfer are both limited to a 12 month period.

Eligibility: Firms with 1 to 9 registered workers (inclusive of the owner) are all eligible to apply to Estimada++. These firms can register up to ten Jordanian and non-Jordanian new eligible workers (inclusive of the owner) under the Estidama++ programme. These workers must earn a wage of less than 500 JOD per month. Priority is given to Syrian refugee workers (where the firm size cap does not apply), migrant workers and female workers.

Two simultaneous and separate randomised interventions aim to encourage eligible firms to participate in Estidama++.

1. Face-to-face intervention: SSC outreach officers will conduct an in-person visit to inform the treatment arm firms about Estidama ++ eligibility and benefits, benefits of social security membership, and explaining the SSC registration process. The outreach officers will hand out a brochure which summarise the information delivered during the visit.
2. SMS intervention: SSC will send a series of SMS messages to the treatment arm firms. The introductory SMS contains the SSC registration number of the firm. The other eight messages will be sent over eight weeks (one message per week). Except the introductory one, all eight messages include a link to the SSC website where registration of workers can be easily completed. The treatment arm firms will be cross-randomized to benefits-framed messages and loss aversion and risk perception messages:
a. The benefits-framed messages emphasise the simplicity of the registration process, highlighting the benefits of the social security package and Estidama++. It also promotes the social norm of social security participation.
b. Messages framed on loss aversion and risk perception emphasise the potential loss and costs that workers and employers may face if they do not participate.

The intervention is expected to take place between February and May 2024.

2024-02-01

2024-04-14

• Registration to social security
• Understanding of social security

• Perception of penalties
• Usage of SSC platform and services

We randomly assign firms to treatment and control groups within each strata.

We randomly assign firms to treatment and control groups, stratifying by social security branch and firm size (very small versus small). Each firm is registered under a specific administrative branch of the SSC. The 'very small firm size' category is defined as a binary variable: it equals one for firms with five or fewer registered workers, and zero for firms with more than five but fewer than ten workers. For further details, please see the pre-analysis plan.

Stratified randomisation conducted using Stata.

Firm

No

Stratified randomisation, no clustering.

Analysis of face-to-face intervention: maximum 1,440 firms using survey data (720 firms in treatment group), and maximum 1600 firms using administrative data (800 firms in treatment group) Analysis of SMS interventions: maximum 43,330 firms using administrative data

Analysis of face-to-face intervention: maximum 1,440 firms using survey data (720 firms in treatment group), and maximum 1600 firms using administrative data (800 firms in treatment group)
Analysis of SMS interventions: maximum 43,330 firms using administrative data (14,444 firms assigned to receive SMS with positive framing treatment, 14,443 firms assigned to receive loss aversion and risk perception framing)

Assuming power of 80% and alpha as 5%, the minimum detectable effect sizes are the following: • Impact of face-to-face intervention on outcomes based on survey: 0.15 SD. • Impact of face-to-face intervention on outcomes based on administrative data: 0.14 SD. s.d. • Impact of one of the SMS interventions on outcomes based on administrative data: 0.03SD