Urban Micro-Insurance Project Evaluation: Pre-Analysis Plan

Last registered on February 22, 2016


Trial Information

General Information

Urban Micro-Insurance Project Evaluation: Pre-Analysis Plan
Initial registration date
March 02, 2015

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
March 02, 2015, 2:07 PM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
February 22, 2016, 7:57 PM EST

Last updated is the most recent time when changes to the trial's registration were published.



Primary Investigator

Princeton University

Other Primary Investigator(s)

PI Affiliation
McGill University
PI Affiliation
Princeton University

Additional Trial Information

Start date
End date
Secondary IDs
This paper describes the analysis plan for a randomized controlled trial evaluating the effect of micro-insurance provision to informal workers in Nairobi, Kenya. In 2011, we worked with the Co-operative Insurance Company (CIC) to provide free health insurance to 300 randomly selected metal workers in Kamukunji Jua Kali. The policy included inpatient and outpatient coverage that applied to the participant’s house- hold. Another randomly selected sample of 300 Jua Kali workers received a one-time unconditional cash transfer equal in value to the insurance policy. This study aims to identify the impact of health micro-insurance vis-`a-vis cash transfers on the economic and psychological well-being of treated individuals. This plan outlines our evaluation questions, outcomes of interest, and a proposed econometric approach.
External Link(s)

Registration Citation

, , Johannes Haushofer and Chaning Jang. 2016. "Urban Micro-Insurance Project Evaluation: Pre-Analysis Plan." AEA RCT Registry. February 22. https://doi.org/10.1257/rct.647
Former Citation
, , Johannes Haushofer and Chaning Jang. 2016. "Urban Micro-Insurance Project Evaluation: Pre-Analysis Plan." AEA RCT Registry. February 22. https://www.socialscienceregistry.org/trials/647/history/6961
Sponsors & Partners

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
Experimental Details


The study worked with a very particular sector in Kenya, the informal sector, commonly known as the “Jua Kali” (literally meaning “under the hot sun”) sector (Orwa 2007). The artisans, vendors, and mechanics in this sector face extreme vulnerability to illness, economic dislocation, and natural disasters. Yet, employment in Jua Kali increased by 5.1 million in 2002, accounting for 74.2 per cent of total non-farm employment, according to Kenyas Central Bureau of Statistics. The Jua Kali sector encompasses small-scale entrepreneurs and workers who lack access to credit, property rights, training, and good working conditions. JKA workers supply goods to local markets using predominantly manual labor and little capital, often making do with handmade tools. Their workshops and stands frequently lack electricity and running water. These workers are organized into various sheds built out of metal sheets that give little protection from the elements. Given the extreme health hazards that they face, Jua Kali workers can stand to benefit from insurance schemes that provide coverage in the event of workplace accidents.

Micro-Health Insurance
Subjects receiving insurance enrolled in the CIC Afya Bora plan, a combined inpatient and outpatient family health insurance policy. These treated households received inpatient benefits of up to KSH 250,000 per family that covered the costs of:
• Hospital accommodation charges for a general ward bed in contracted hospitals • Doctor and healthcare professional fees
• Prescribed routine lab tests
• X-ray and ultrasound tests
• ICU, HDU, and theatre charges
• Prescribed medicines, dressings, and internal surgical appliances
• Routing diagnostic lab tests
• Day care surgery
• Maternity including non-elective caesarean section with 6 mo. waiting period
• Chronic and pre-existing conditions up to KSH 75,000
Households also received outpatient benefits of up to KSH 50,000 per family that
• Routine outpatient consultation
• Diagnostic laboratory and radiology services
• Prescribed medicine and dressings
• HIV/AIDS related conditions and prescribed ARVs
• Routine immunizations
• Routine prenatal check ups
• Postnatal care up to six weeks after delivery
• Pre-existing and chronic conditions up to KSH 20,000 • Outpatient oncology
• Psychiatry and psychotherapy
Beneficiaries paid KSH 100 for each outpatient visit. Both covers included chronic and pre-existing conditions, including HIV/AIDS but excluded treatment outside Kenya, cosmetic treatment, treatment by non-qualified persons, infertility, self-inflicted injury, ex- perimental treatment, and dental treatment unless occassioned by accidental injury. Ben- eficiaries could access these benefits through CIC’s network of providers that included 26 mission and faith based hospitals in Nairobi.
The plan provided benefits to principals and spouses under 72 years old and children dependents younger than 25 years with proof of enrollment in school or college. Subjects were enrolled in the Afya Bora plan free of charge for one year, a value of KSH 12,745 for the principal, spouse and up to five dependents. Each additional child dependent increased the annual premium by KSH 2,000 per child. The project fully reimbursed households for the base cost and any added premium.

Unconditional Cash Transfer
Subjects in the second treatment group received an unconditional cash transfer equal to the net value of the annual premium they would have had to pay had they enrolled in the CIC Afya Bora scheme. The transfer was delivered to recipients electronically using the M-Pesa mobile money service. M-Pesa is a mobile money system offered by Safaricom, the largest Kenyan mobile phone operator. Using M-Pesa requires a registered SIM card and a valid Kenyan national ID card. The project transfered the money from Innovations for Poverty Action Kenya’s (IPA-K) M-Pesa account to that of the recipient. To facilitate the transfers, we encouraged recipients to sign up for M-Pesa and helped them obtain, where necessary, all of the requirements for registration. The cash was transferred to the registered SIM card wirelessly and the recipient could withdraw the balance at an M-Pesa agent by putting the SIM card into the agents cell phone or by using their own phone.
Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)
Please see attached pre-analysis plan.
Below are a list of variables of interest from the pre-analysis plan.

6.1 Assets
1. Cell phone
2. Sofa or chairs
3. Piped water
4. Clock/watch
5. Bicycle
6. Radio, tape, or CD player 7. Battery
8. Generator
9. Motorcycle
10. Car/Truck
11. Solar panel
12. Television or computer
13. Farming tools
14. Wheelbarrow
15. Cart
16. Kerosene stove
17. Refrigerator
18. Renting or owning a house
19. Moved to different house
20. House has electricity
21. Total value of assets*
22. Asset Index*: Weighted standardized average of 1 - 17
6.2 Consumption
1. House rent
2. House mortgage
3. Drinks (non-alcoholic)
4. Airtime, Internet
5. Cigarettes/alcohol
6. Restaurant/prepared meals 7. Travel, transport, and hotels 8. Gambling
9. Clothing
10. School fees and supplies
11. Medical expenses
12. Fixing fire damage
13. Fixing water damage
14. Work materials
15. Religious expenses
16. Social expenses
17. Gifts to friends
18. Electricity
19. Water
20. Domestic Staff
21. Insurance
22. Bride price
23. Fuel
24. Total annual expenditure*
25. Health expenditure* (Medical expenses, Health insurance)
26. Temptation goods expenditure* (Gambling, Alcohol/Cigarettes)
27. Social expenditure* (Restaurant/prepared meals, Religious expenses, Social expenses, Gifts to friends, Bride price)
6.3 Borrowing and Savings
1. Have any loans
2. Total amount borrowed*
3. Ability to repay loans*
4. Remittances received in past month
5. Remittances sent in past month
6. Amount currently saved
7. Amount saved each month in social group* 8. How secure do savings make you feel?
9. Can savings cover health expenses?
6.4 Health
injury in the past month injury is work-related
of household sick or injured* of children sick or injured*
1. Sickness or
2. Sickness or
3. Proportion
4. Proportion
5. Number of
6. Days of work/school missed due to illness/injury in the past month
chilren passed away within last year
Healthcare use
1. Child vaccination
2. Proportion of children vaccinated*
3. Child preventative care check-ups
4. Consulted with health care provider for illness/injury
5. Total treatment costs associated with illness/injury
6. Number of nights hospitalized over the past year
7. Total cost of hospitalization in the past year
8. Number of nights in the past year where member should have been hospitalized but wasn’t
9. Ability to pay for medical treatment
6.5 Labor
1. Will leave JKA
2. Will change occupation within JKA
3. Will move to a riskier occupation*
4. Average weekly income in the past year 5. Last week’s income
6. Predicted weekly income next week
7. Average number of hours worked per day 8. Average number of days worked per week 9. Involved in production, sales, or both
10. Self-employment
11. Average goods produced per day
12. Goods produce per day last week
13. Productivity per day*
14. Attended school in the past year
15. Took formal training course in the past year
16. Took informal training course in the past year
17. Shed leader
18. Level of trust in shed members
19. Perceived risk of own job
20. Labor Mobility Index*: Weighted standardized average of 1 - 3
21. Productivity Index*: Weighted standardized average of 4 - 8, 11 - 13
6.6 Insurance
1. Trust in insurance companies
2. Likelihood of keeping CIC insurance
3. Owning fire insurance
4. Owning inpatient insurance
5. Owning outpatient insurance
6. Owning life insurance
7. Owning accident insurance
8. Willingness to pay for fire insurance
9. Willingness to pay for inpatient insurance
10. Willingness to pay for outpatient insurance
11. Willingness to pay for outpatient insurance with co-pay
12. Willingness to pay for life insurance
13. Willingness to pay for critical illness insurance
14. Number of times used fire insurance
15. Number of times used inpatient insurance
16. Number of times used outpatient insurance
17. Number of times used outpatient insurance with co-pay
18. Number of times used life insurance
19. Number of times used critical illness insurance
20. Insurance Ownership Index*: Weighted standardized average of 3 - 7 21. Insurance WTP Index*: Weighted standardized average of 8 - 13
22. Insurance Usage Index*: Weighted standardized average of 14- 19
6.7 Psychological & Neurobiological Welfare
1. Summary of worry incidents*
(a) Worry over health problems
(b) Worry over accidents and disasters
(c) Worry over problems in the workplace􏰁 (d) Worry over finding work􏰁
(e) Worry over losing employment􏰁
(f) Worry over having too much work to do􏰁
(g) Worry over having enough money for basic needs􏰁 2. Cortisol level
3. Cytokine level
4. Perceived Stress Scale score
5. Locus of Control score
6. Scheier Optimism score
7. Self-Esteem Scale score
8. CES-D
9. World Value Survey happiness
10. World Value Survey satisfaction
11. Self-Reported Welfare Index*: Weighted standardized average of 4 - 10 12. Neurobiological Welfare Index*: Weighted standardized average of 2 - 3
6.8 Preferences
1. Impatience
2. Decreasing impatience
3. Risk aversion
4. Other-regarding preferences
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Please see attached pre-analysis plan.

The project employed a randomized evaluation strategy in order to identify and evaluate the causal effect of providing health insurance to economic and psychological measures of welfare. We studied a randomly selected sample of metalworkers of the Kamukunji Jua Kali Association (JKA) in Nairobi, an organization comprised of an estimated 4,000 Jua Kali workers. Only adult JKA members who work in an area of Kamukunji Jua Kali that makes him or her eligible for voting rights with the JKA were eligible to participate in the research. We ran a preliminary survey including 1,392 JKA members between August 2008 and December 2010. We also ran a pilot study between Feburary 2010 and December 2010 to test the questionnaire and identify potential difficulties in the scale-up of the project.
After the census, the subjects were randomly stratified into three income groups ac- cording to household weekly income. 313 subjects with a weekly income greater than KSH 4,000 comprised the high income group. 300 subjcts with a weekly income between KSH 2,000 and KSH 4,000 comprised the middle income group. 242 subjects with a weekly income under KSH 2,000 comprised the low income group. We randomly selected subjects into one of the two treatment arms or the comparison group within each income strata.
Experimental Design Details
Randomization Method
Please see attached pre-analysis plan.
Randomization Unit
Please see attached pre-analysis plan.
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
900 individuals
Sample size: planned number of observations
900 individuals
Sample size (or number of clusters) by treatment arms
300 control, 300 cash, 300 health insurance.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number
IRB Name
IPA Kenya
IRB Approval Date
IRB Approval Number
IRB Name
McGill University
IRB Approval Date
IRB Approval Number
Analysis Plan

Analysis Plan Documents

Urban Micro-Insurance Project Pre-Analysis Plan - Addendum

MD5: 73dafe0b9efe81eedc61131ab5717508

SHA1: 04664a389a28af12b2e7089e94fbea55686cbd90

Uploaded At: February 22, 2016

Urban Micro-Insurance Project Pre-Analysis Plan

MD5: 46218e27a58329f34641a1deeee1078a

SHA1: 3f366f9d844622b4a1f0b6d03594db65ef3b6463

Uploaded At: March 02, 2015


Post Trial Information

Study Withdrawal

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information


Is the intervention completed?
Data Collection Complete
Data Publication

Data Publication

Is public data available?

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials