State Dependence in Poverty: The Case of Turkey

ABSTRACT

Despite poverty alleviation efforts, almost a quarter of households live below the poverty line in Turkey. This article aims to examine the dynamics of poverty focusing on poverty persistence in Turkey, utilizing Income and Living Conditions panel data belonging to 2010–2013. A random effects dynamic panel probit model has been employed. In order to tackle the initial values problem Heckman’s reduced form approximation is utilized. Empirical results indicate that gender, educational attainment, employment type, and household structure have statistically significant impact on the probability of being poor. Besides, experiencing poverty has a positive impact on future poverty likelihood, signalling state dependence.

KEY WORDS:

• poverty
• random effects dynamic probit model
• state dependence
• Turkey

Notes

1. Please see Yildirim, Ozdemir, and Sezgin (2014) for the qualitative evaluation of Conditional Cash Transfer Program in Turkey.

2. Please see Yildirim and Dal (2016) for a summary of Social Assistance System in Turkey.

3. OECD socx social expenditure database: https://stats.oecd.org/Index.aspx?DataSetCode=SOCX_AGG.

4. The model considers the issue of time-invariant unobserved heterogeneity. As the time period is short, ignoring time-variant individual effects is not expected to create problems.

5. Stewart (2006) has provided a STATA command, redprob, for estimating Heckman’s estimator, which has been employed in this study to control for initial conditions.

6. For details of this approach please see Stewart (2007).

Additional information

Funding

This research was financially supported by the Scientific and Technological Research Council of Turkey (TUBITAK) project number 115K574.