Children and Familial Economic Welfare: The Effect of Income on Child Development

1. Introduction

Research in Canada and elsewhere has shown a link between economic disadvantage and a broad range of poor child developmental outcomes (Blau, 1999; Ross and Roberts, 1999; Duncan et al., 1998; McLoyd, 1998; Duncan and Brooks-Gunn, 1997; Mayer, 1997; Duncan et al., 1994). Lack of immediate as well as social or communal resources all contribute to the overall economic disadvantages faced by some families with children relative to other families. On the one hand, children in families with greater financial and economic resources generally have more secure living conditions and attachments, in addition to greater access to social, health, educational, and recreational opportunities - important factors shaping positive and healthy development. Children from low-income families, on the other hand, are at significant social, physical and emotional risk, especially for behavioural and emotional problems, and poor academic performance (Duncan et al., 1998 and Duncan et al., 1994). Despite this research, however, the exact nature and strength of the relationship between a family's economic circumstances and children's developmental outcomes is still open to debate (Blau, 1999; Mayer, 1997). Research regarding familial changes such as divorce, separations and remarriage, for example, suggests that these changes also have a direct effect on a family's economic circumstances as well as on children's outcomes (Jekielek et al., 1998; Thomson et al., 1994).

While it is the case that the exact relationship is still being debated, many researchers do contend that a child's welfare and development is connected in part to their family's level of economic welfare. A family's economic welfare in this sense encompasses its general level of income, employment related benefits, type of work and employment of those earners in the family, level and availability of services from the private and public sector, as well as the general economic conditions in the community. Indeed, the very different causes of change to a family's economic conditions open the possibility of very different policy initiatives to address the experiences of children as they development. Our aim in this project will be to examine only one aspect of the economic welfare of a family, namely the impact that income has on children's development in Canada.

Previous studies in Canada have been largely based on cross-sectional data (Ross and Roberts, 1999; Lefebvre and Merrigan, 1998; Lipman and Offord, 1997). This work has been useful in identifying the specific linear and non-linear effects of income for children, as well as those factors which serve as either protective mechanisms or enhance children's developmental opportunities. But much remains to be done in understanding the effect of income and the experience of poverty (CCSD, Key Informant Study, 1999) on children's later development. We need to look at not only the level of income, but the dynamics or flows of income as well.

New sources of longitudinal survey information on children in Canada provide a starting point to allow a determination of the extent to which the level and stability of income is an important factor in shaping the life chances of children, and possible areas for public policy intervention. Unfortunately these surveys are still very much in their infancy. As more years and a greater number of cycles of information become available we will be able to obtain a more detailed picture of the relationship between a family's economic conditions and their children's development. Our study, therefore, provides only an initial assessment of the evidence of the effect of income and child development. As well, given the importance of other economic factors, we provide a preliminary examination of the effect of labour market changes (as measured by number of family earners) and changes in family composition (as measured by changes to two parent and lone parent status) on the behavioural outcomes of children. However, the main focus of our analysis will be on the effect of income, its level and stability on the development of children.

Governments at both the provincial and federal levels are struggling with the best way to ensure that all children in Canada realize their full potential. Knowing what to do is affected in part by an understanding of the experience and impact of income and changes in family income on children's development. To this end, and bearing in mind the limitations of the survey data, we address the following questions:

How much income fluctuation and of what magnitude do children's families experience year to year?
What proportion of children spend sometime in lower income situations? How long do children spend living in poverty? What proportion leave or enter lower income situations?
How important is income in affecting children's outcomes as measured by cognitive and behavioural measures?
Does income have a different effect on children at different ages or stages of development?
What proportion of these income fluctuations are related to labour market changes and what proportion are related to changes in familial structure?

We answer these questions in our study, which is divided into ten sections, with data drawn from the share file of the first two cycles (1994-1995 and 1996-1997) of the National Longitudinal Survey of Children and Youth (NLSCY). We focus on the cohort of children aged from 2 to 13 years old in 1996 (ages 0 to 11 in 1994), yielding a sample of 15,266 children to investigate the research questions.¹ Section one introduces the study. The second section reviews the relevant literature regarding the effect that income security, change in family structure, and changes in familial employment, have on children's outcomes. It is followed by a discussion of the methodological issues and hypotheses to be tested. Our fourth section is an initial analysis of the economic security of families and how it affects children's outcomes. We measure a family's economic circumstances by observing changes in income, family structure, number of earners and child outcomes across the two cycles of the NLSCY. We then use cross-tabular analysis to observe the correlation between families' changing economic circumstances and children's developmental outcomes. Our results show the distribution of household income in each of the two cycles to be generally the same. This equivalence, however, hides broad income shifts that are correlated with changes in family structure and number of earners.

The analysis then proceeds to investigate, in our fifth section, the effect of income on child outcomes employing a typical reduced form OLS regression model (Blau, 1999; Mayer, 1997). The results of this analysis demonstrate that the effect of income on children's behavioural and cognitive outcomes is for the most part significant, even after the application of controls. Nevertheless, we find that the effect of income is relatively small, which concurs with much of the previous research. Based on evidence in the research literature that the effect of income may be working through the home environment of the child (Blau, 1999; Jekielek et al., 1998; Schiamberg, 1991), we construct our own home environment index. Our index is only a very preliminary and much pared down proxy of the psychometric scale measure used in previous research in America. We observe that income is related to the home environment measure with children from affluent families tending to have slightly better and more stimulating home environments. This suggests that income plays a part in affecting children's outcomes through the home environment.

Following the data analysis our findings are discussed in more detail in section six. This leads to section seven where the policy implications of the study are addressed. In section eight, we conclude by making some recommendations regarding directions for further research in the area of economic security, including income and poverty dynamics, and into how useful it may be to try and obtain a home environment measure. The ninth section contains the appendices and tables of the study. Our tenth and last section contains the references.

¹The share file is a nationally representative file of respondents to the NLSCY who permitted the use of their responses by Human Resources Development Canada. It represents approximately 95 per cent of respondents to the NLSCY.

Contents

Children and Familial Economic Welfare: The Effect of Income on Child Development - April 2001

1. Introduction