Interactions Between Policy Effects, Population Characteristics and the Tax-Benefit System: An Illustration Using Child Poverty and Child Related Policies in Romania and the Czech Republic

Soc Indic Res (2016) 128:1365–1385 DOI 10.1007/s11205-015-1083-6 Interactions Between Policy Effects, Population Characteristics and the Tax-Benefit System: An Illustration Using Child Poverty and Child Related Policies in Romania and the Czech Republic Silvia Avram1 • Eva Militaru2 Accepted: 21 August 2015 / Published online: 5 May 2016  The Author(s) 2016. This article is published with open access at Springerlink.com Abstract We investigate the impact of the Romanian and Czech family policy systems on the poverty risk of families with children. We focus on separating out the effects of policy design itself and size of benefits from the interaction between policies and population characteristics. We find that interactions between population characteristics, the wider tax benefit system and child related policies are pervasive and large. Both population characteristics and the wider tax-benefit environment can dramatically alter the antipoverty effect of a given set of policies. Keywords Child poverty  Child benefits  Microsimulation  Policy interactions  Population interactions 1 Introduction The past 20 years have witnessed prominent policy initiatives to tackle child poverty both at the European and national levels (for example, the Lisbon strategy or the Labour government pledge to halve child poverty in the UK by 2020). However, despite these efforts, child poverty rates have remained stubbornly high. Even more worryingly, they have increased in some countries especially in comparison with overall poverty rates (Oxley et al. 2000; Van Mechelen and Bradshaw 2013). For example, between 2005 and 2012, poverty among children in the 27 Member States has broadly remained stable around 28 % whereas poverty among the population as a whole fell from 26 to 25 % (EUROSTAT). & Silvia Avram 1 ISER, University of Essex, Wivenhoe Park, Colchester CO43SQ, UK 2 National Research Institute for Labour and Social Protection, 6-8 Povernei Street, Sector 1, 010643 Bucharest, Romania 123 1366 S. Avram, E. Militaru A large body of scholarly work has linked poverty, and low income in general, to deleterious consequences on child developmental trajectories and educational attainment (Black et al. 2000; Engle and Black 2008; Najman et al. 2009; Petterson and Burke Albers 2001), health status (Aber et al. 1997; Case et al. 2002), as well as adulthood outcomes (Duncan et al. 1998, 2010). Given the consequences of material deprivation both on current well-being and future capability and the fact that children generally have little control over what economic resources are available to them, there is overwhelming agreement that child poverty is an area necessitating public intervention. To mitigate child poverty, governments can resort, among other tools, to various forms of income support and child contingent transfers. Previous scholarly work has found considerable evidence that child contingent transfers do have a substantial effect on child poverty outcomes, with typically large cross-national variation in policy effects (Matsaganis et al. 2007; Barrientos and DeJong 2006; Bradshaw 2006; Immervoll et al. 2000; Whiteford and Adema 2007). These studies usually use either pre-transfer post-transfer comparisons or a microsimulation-based approach and attribute any differences in observed poverty or inequality indicators to the policy package they investigate. One aspect left unaddressed in these studies is the extent to which policy effects are shaped by ‘outside’ factors, i.e. population characteristics and/or the wider taxbenefit system in which they operate. Although these studies generally acknowledge the existence of interactions of various sorts and their potential in shaping the impact of family transfers, they fail to explicitly investigate these issues. As a result, there is little evidence on the sensitivity of estimated policy effects to variation in the population profile and the design of other social and fiscal instruments that are present. For example, can these factors alter the ranking of policy instruments with similar objectives? These issues are all the more important as the European Union (EU) has launched various benchmarking exercises that essentially rely on comparisons between countries with potentially very different demographic, labour market and tax-benefit institutions. This paper seeks to bridge this gap and contribute to the understanding of the role of interactions between child contingent policies, population characteristics and the wider tax benefit system in shaping the impact of the former on child poverty. By interactions, we mean that the magnitude of the policy effect is itself contingent on other factors, in particular population characteristics and/or the architecture of the wider tax-benefit system. To this end, we take Romania and the Czech Republic as case studies and examine the reduction in child poverty effected by three family transfers and one tax concession (see Table 1). Romania is a country with high levels of child poverty where the support package available to families with children has been found to be not very effective (TARKI 2010). In contrast, the Czech Republic registers low overall and child poverty rates which have been found to be at least partly the result of generous income support (TARKI 2010). Using microsimulation techniques, we examine to what extent these results are driven by the characteristics of the child-related policy instruments themselves as opposed to being the product of the ‘fit’ between these instruments, other income support measures available to families with children and population features. More specifically, we compute the direct, first-order effect of both the Romanian and the Czech child policy package on relative poverty, while varying the underlying population characteristics and the wider tax-benefit system. Following Salanauskaite and Verbist (2013), we also distinguish between instrument generosity and instrument design in measuring the impact of a given child policy package. The rest of the paper proceeds as follows. Section 2 reviews the existing literature on the links between child related transfers and child poverty. Section 3 describes the Romanian and Czech policies we consider in this exercise. Section 4 describes the data and 123 Interactions Between Policy Effects, Population… 1367 Table 1 Policy instruments included in the child package Policy Eligibility Amounts % Of children in families receivinga Allowance for new born children and the outfit for new born children Universal entitlement for all new-borns Lump sum of approx. 354 RON 6 2 Universal child benefit Age \18 or in high school Per month/200 RON for children under 2; 25 RON for children 2 and older 100 7 Means-tested family benefits Means-tested; monthly income \176 RON per person; children are persons \16 or \18 and with family income\50 RON/month Between 36 and 52 RON/month, depending on th (...truncated)