Cash transfers, conditional or unconditional, are now a major policy instrument in developing countries, with over two dozen countries implementing some form of cash support payments to poor families. The Government of Kenya’s Cash Transfer for Orphans & Vulnerable Children (CT-OVC) is the country’s flagship social protection program, reaching over 150,000 ultra-poor households nationwide. The objective of the program is to provide support to families with orphans and vulnerable children in order to facilitate their development.

This paper assesses the impact of the CT-OVC on a range of fertility related outcomes including age at first sex, early marriage, pregnancy, forced sex, and the number of unprotected sex acts in the last three months. The data are derived from the second follow-up household survey of the impact evaluation of the CT-OVC among adolescents age 15-21. The evaluation design is a cluster randomized longitudinal study with baseline collected in 2007, first follow-up in 2009 and second follow-up in 2011. Households were surveyed in seven sites across the country, and the 2011 panel is composed of 1900 households, two-thirds of whom are in the intervention group and the remaining one-third in the control group.  There are 3000 adolescents age 15-21 in the sample, roughly half of whom are female.

The outcomes of interest in this paper were collected in the 2011 round only so we limit the sample to individuals whose first sex act occurred after the intervention began in 2007. We use cross-sectional multivariate analysis to test whether the outcomes of interest differ by intervention status. We see strong impacts of the program on delayed sexual debut for both males and females. For females we find evidence of a positive impact of the program on delayed marriage and pregnancy. We hypothesize that these effects may be working through increased schooling, exposure to different peers or psycho-social status (depression). WE have collected data on these three factors and will test these hypothesize using standard tests for mediation in order to understand how the program has affected the outcomes of interest.

To our knowledge this is the first paper to test the impact of a large scale government program on  sexual debut, teenage pregnancy, and marriage.