How data analytics can help us to end modern slavery: Examples from the Philippines
Kerry Bruce, DrPH @kbruce2001 @SocImpactInc Greg van der Wink, Ph.D @Novametrics Direct Measurement of Modern Slavery
• Difficult to measure • Difficult to define • Ethical Issues • Safety Issues What about Indirect Measurement?
• Using techniques from environmental, geophysical sciences • Is there a way to “characterize” the ecosystem of modern slavery? • Can we use indirect data to point us to where vulnerabilities are? Hypotheses – vulnerabilities to trafficking
1. High un/underemployment would mean high prevalence of OFWs. 2. People are becoming OFWs to make more money 3. Regions with high numbers of OFWs will have higher numbers of victims Potentially Exploitive Limited Societal Vulnerable Population Industry (low skill, low pay) Safeguards (weak capacity) • Manufacturing /Factory • Economic (poverty) • Rule of Law (labor laws, (Clothing Processing) • Education (gender enforcement, • Domestic Service differences, access, prosecution) (housekeeping, OFW) attainment) • Social Safety Nets (old • Health (gender age pensions, health differences, access, age coverage, support for at first sex) disabled and victims) • Sociocultural (race, • Institutional Systems ethnicity, gender (access to money, access inequality) to land, OFW system) • Political (Immigration • Risk Perception policy, conflict, power of (knowledge of rights, local government) family influence) Vulnerable Population Modern Slavery
Limited Exploitive Societal Industry Safeguards
Prevalence of OFWs does not increase with increasing unemployment or underemployment rates
Unemployment Underemployment 70 70
60 60
50 50
40 40
30 30
20 20
OFWs / 1000 Working Age People Age Working 1000/ OFWs OFWs / 1000 Working Age People Age Working 1000/ OFWs 10 10
0 0 3 4 5 6 7 8 9 10 11 0 5 10 15 20 25 30 35 40 Unemployment Rate Underemployment Rate
Source: Philippines Statistics Authority 250
200 Remittances, Skill Levels and Gender
150 Potential priority target for interventions.
100 Remittance per OFW pesos) (x1000 OFW per Remittance
50
51% 0
Officials and Technicians and Clerks and Trade and Laborers and Professionals Associate Professionals Service Workers Machine Workers Unskilled Workers Skill level (high low)
Over 50% of the female OFWs are working as unskilled workers providing remittances of less than $750/ year. Which regions have high prevalence of OFWs and Victims?
250 70
60 200
50
150 40
30
100 OFWs RVs/1000
OFWs/1000 Working Age People Age Working OFWs/1000 20
50 10
0 0 NCR CAR I-ILO II-CAV III-CLU IVA-CAL IVB-MIM V-BIC VI-WVS VII-CVS VIII-EVS IX-ZAP X-NMI XI-DAV XII-SOC XIII-CGA
OFWs (Female) OFWs (Male) Victims (Female) Victims (Male)
Note: The Visayan Forum Foundation has regional centers located in Metro Manila (NCR), Batangas (IVA) , Bacolod (NIR), Davao (XI), and Sorsogon (V). Source: Visayan Forum Foundation, Inc., International Organization for Migration, Philippines Statistics Authority Vulnerability vs. OFW and RV prevalence
70 70
60 60
50 50
40 40
30 30 RVs / 1000 OFWs / 1000 RVs 20 20
10 10 OFWs / 1000 Working Age People Age Working 1000 / OFWs
0 0 -40 -20 0 20 40 60 80 100 120 140 160
Most Vulnerable Least Vulnerable
Sources: Visayan Forum Foundation, Inc., International Organization for Migration, Philippines Statistics Authority, Philippines Department of Education, UN Office for the Coordination of Humanitarian Affairs, National Statistics Coordination Board, Food Nutrition Research Institute How can evaluators use this approach?
• Recommend using secondary data analysis/data analytics BEFORE evaluation – Better understand what is happening – Better able to target data collection in the right areas – Create a descriptive and predictive model • Use primary data collection to feed back in to the model to move towards prescriptive solutions • Keep building the model as you go along socialimpact.com 2300 Clarendon Blvd. #1000, Arlington, VA 22201 703.465.1884