Skip to main navigation Skip to search Skip to main content

Kazakhstan’s Workfare Program: A Multidimensional Evaluation

Peter Howie, Indira Davletova, Indira Makhazhan

Research output: Working paper

Abstract

Workfare programs (WPs) in low- and middle-income countries have become a primary instrument to fight poverty associated with temporary economic shocks or to create a pathway out of chronic poverty. In this paper, we examine a WP in Kazakhstan. We do so by executing a cost effectiveness analysis (CEA) using Ravallion’s (1999) framework for rapid appraisal of WPs in low- and middle-income countries and performing a case study analysis which investigates how the implementation process unfolded and what factors shaped it most. We find that the cost per unit of current benefit of Kazakhstan’s WP is significantly higher than both Ravallion’s (1999) value and Adato and Haddad’s (2001) average value. This high value is, in part, the result of benefit ‘leakage’ to the non-poor, high levels of informal labor, and means-testing applicants. The qualitative case study indicates that overly bureaucrat requirements of official residency and household asset documents, poor staff training because of high turnover, staff not following standardized procedures because of low wages and corruption, and central government KPIs resulted in “creaming” of the best applicants and improper blending with other active labor market policies. Our findings provide important lessons for policy makers planning to implement their own WP.
Original languageEnglish
PublisherJournal of Social Policy
Publication statusSubmitted - Jul 16 2021

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 16 - Peace, Justice and Strong Institutions
    SDG 16 Peace, Justice and Strong Institutions

Keywords

  • Workfare; cost-effectiveness analysis; implementation assessment; Kazakhstan

Fingerprint

Dive into the research topics of 'Kazakhstan’s Workfare Program: A Multidimensional Evaluation'. Together they form a unique fingerprint.

Cite this