Abstracts
The Employment and Economic Well Being of Working-Age Men with Disabilities: Comparing Outcomes in Australia, Germany, and Great Britain with the United States
Richard V. Burkhauser, Maximilian D. Schmeiser, Mathis Schroeder
This paper using Cross-National Equivalent File data traces the relative employment of and use of government transfers by working-age men with disabilities in three modern industrial economies with quite different social institutions (Australia, Great Britain, and Germany) and compares them with a similar population in the United States using Current Population Survey data.
Our results suggest that the long run decline in the relative employment and increased dependence on government transfers of United States working-age men with disabilities since at least the 1990s has not occurred in Germany—a country that relies much more heavily on a work first strategy to maintain its disabled workers in the labor force—or in Great Britain and Australia—countries that rely much more heavily on government transfer programs to provide income to their working-age men than does the United States.
A Comparative Analysis of the Nativity Wealth Gap
Deborah Cobb-Clark, Thomas Bauer, Vincent Hildebrand, Mathias Sinning
This paper investigates the sources of the gap in the relative wealth position of immigrant households residing in Australia , Germany and the United States. Our results indicate that in Germany and the United States wealth differentials are largely the result of disparity in the educational attainment and demographic composition of the native and immigrant populations, while income differentials are relatively unimportant in understanding the nativity wealth gap. In contrast, the relatively small wealth gap between Australian- and foreign-born households cannot be explained by either the distribution of income or immigrants’ characteristics. In particular, immigrants to Australia are older, have fewer children and are more educated than their native-born counterparts which serves to narrow rather than widen the nativity wealth gap.
The Impact of Children on Australian Women's and Men's Superannuation
Nick Parr, Shauna Ferris, Stephane Mahuteau
Using data from Wave 2 of the Household Income and Labour Dynamics in Australia (HILDA) Survey, this paper examines how superannuation savings by women and men vary according to the numbers of children they have. The results show that for women there is a clear inverse relationship between the value of superannuation and the number of children they have. Moreover the inverse relationship between a woman’s value of superannuation and her number of children persists after controlling for an extensive range of variables which may affect both her number of children and her superannuation. The analysis also shows that level of education, migrant status, being an employer or self employed, marital status, age and sex are significantly related to an individuals’ level of superannuation. Suggestions for further research and the implications of the results for Australia’s public debate are discussed.
Assessing Change in the Distribution of Domestic Labour over the Lifecourse
Janeen Baxter, Michele Hayne and Belinda Hewitt
It is well-known that pathways through the lifecourse have changed in recent years. People are marrying later, having fewer children, living together in cohabiting relationships, separating and divorcing more frequently. These changes have consequences for understanding the organisation of domestic work. Although much previous work on domestic labour has focused on married couples, it is becoming increasingly clear that we need to consider how housework patterns vary at different stages of the lifecourse and in different kinds of households. This is important not just because we need to acknowledge household diversity, but because research has shown that previous relationship experiences will affect the ways in which individuals and couples organise domestic labour in their current households. In this paper we build on our earlier work in which we examined the effect of different pathways into marriage on housework patterns during marriage. We extend this earlier work in three main ways. First we consider different pathways into and out of marriage and the effect of these pathways on housework hours across different household types for both men and women. Second we develop a new equivalised measure of housework distribution that enables us to examine changes in the relative distribution of housework with changes in relationships and household composition, including movements from and to lone person households. Third we use 5 waves of HILDA data to examine varying pathways into and out of relationships and the affect of these pathways and changes in the composition of the household on the domestic division of labour over the lifecourse.
Do Perceived and Actual Inequity in Housework Shares Predict Marital Survival?
Pooja Sawrikar
Despite the large-scale entry of women into the paid workforce there has been little corresponding movement by men into domestic labour. Given the rising rates of divorce and the falling rates of birth in Australia, it remains to be seen if (dis)satisfaction with the division of household labour predicts marital survival and the number of children women have (net of other predictors). The purpose of this study is to exploit the longitudinal nature of the HILDA dataset across Waves 1 to 4, using logistic regression, and test the hypotheses that the wife’s and husband’s (i) perceived fairness in the share of housework, (ii) actual share of hours per week spent on household labour, and (iii) actual number of hours per week spent of household labour significantly predicts the odds of two events occurring – marital dissolution (from Wave 1 to 4) and whether couples with one child at Wave 1 will have had more children by Wave 4. The first hypothesis in relation to marital survival is based on a sample of 2500 married couples at Wave 1, and the second hypothesis in relation to the birth of at least a second child is based on a sample of 114 heterosexual couples. The control variables across both analyses span several variables including: satisfaction with work-family balance, gender ideology, relationship satisfaction, importance of religion, parent’s marital history, cohabitation history, women’s share of household income, education, employment status, ethnicity, and SEIFA index of relative dis/advantage. Cross-sectional results on the relationship between the main predictor variables will also be presented. The results showed that perceived fairness in the share of housework and the number of hours that men contribute to domestic labour each week are significant predictors of marital survival and that women’s share of housework hours per week significantly predicts whether she will have more than one child. These results will be discussed conceptually in relation to gender equity, and methodologically in relation to data quality cross-sectionally and longitudinally.
Long Work Hours: Volunteers and Conscripts
Robert Drago, Mark Wooden and David Black
Panel data from Australia are used to study the prevalence of work hours mismatch among long hours workers and how that mismatch persists and changes over time, and what factors are associated with these changes. Both static and dynamic multinomial logit models are estimated, with the dependent variables distinguishing long hours workers from other workers, and within the former, between “volunteers”, who prefer long hours, and “conscripts”, who do not. Particular attention is paid to choice theoretic reasons for long hours, the role of bargaining power, and norms around consumerism and ideal workers. Consistent with earlier mismatch studies, the results suggest that labour markets function sluggishly at best, creating a large group of persistent long hours conscripts. Relatedly, choice theoretic considerations appear to be of little importance, although employed fathers in traditional “breadwinner” families are more likely to be long hours volunteers. Bargaining power, as indicated by union membership, public sector or self employment provides a superior explanation for the phenomena. Norms around work and spend and around ideal workers provide the strongest explanation, with high levels of debt and ideal worker status tending to drive and sustain long hours conscript status.
Financial Incentives Affecting the Retirement Decisions of Mature Age Australians
Diana Warren and Umut Oguzoglu
In Australia, labour force participation among older people, particularly men over the age of 55, has been declining over the last 30 years. Previous research has found that in many OECD countries, the retirement income system actually provides incentives for older workers to retire early rather than remain in the work force. We use data from the first five waves of the Household, Income and Labour Dynamics in Australia (HILDA) survey to identify any financial incentives present in the Australian retirement income system on retirement decisions of men and women aged between 54 and 64 in 2001. As the superannuation guarantee only came into effect in 1992, the majority of people in this age group would not have built up substantial superannuation savings, and most will rely at least partially on the age pension in retirement.
A unique feature of the HILDA data is that detailed information on wealth, superannuation and income support payments is provided. Following Gruber & Wise (2004), we model retirement behaviour where individuals retire in the period that the present value of their lifetime retirement income is maximised. We also utilise an augmented model that considers the trade-off between utility drawn from leisure and utility drawn from labour income.
Preliminary findings from our econometric model suggest that for men the Australian retirement system provides incentives to retire early, while for women financial incentives are less significant, as the factors that influence women’s retirement behaviour are more commonly found to be health and family related, rather than financial incentives.
Modelling the workers of tomorrow: the APPSIM dynamic microsimulation model
Ms Marcia Keegan, PhD Student, National Centre for Social and Economic Modelling, The University of Canberra
As the recently released Intergenerational Report makes clear, population ageing is expected to create severe fiscal pressures for governments. With 13 Commonwealth agencies as research partners, NATSEM is currently constructing the Australian Population and Policy Simulation Model (APPSIM), to simulate our likely social and economic futures and the future distributional impact of policy changes. The APPSIM model takes the 1996 Census one per cent sample file as its base data and then ages the individuals within the sample, year by year, to about 2050.This paper provides an overview of APPSIM and then describes in more detail the labour force participation module for APPSIM and the methodology and data sources being used to develop it. The purpose of the labour force module within APPSIM is to project the distribution of labour force participation for the next 50 years.
The primary source of data for the module is HILDA, the Household, Income and Labour Dynamics in Australia Survey.
The five waves of HILDA are used to determine the likelihood that individuals within APPSIM will change their labour force status from one year to the next, given their age, sex, education, marital status, children, health and working history. Since the previous year’s labour force status is one of the strongest predictors of the current year’s labour force status, the longitudinal nature of HILDA is invaluable to the development of this model.
The paper contains a brief summary of patterns and projections of Australian labour force participation. It then discusses the methodology used in the development of the equations to predict labour force participation. It goes on to describe how the module projects the four possible labour force statuses of full time work, part time work, unemployed and not in the labour force. Finally, the paper discusses the alignment of the model to user-definable aggregate projections of labour force participation.
Job Satisfaction and Gender: Evidence from Australia
Temesgen Kifle and Parvinder Kler
This paper investigates six different aspects of job satisfaction (pay, job security, work, hours, flexibility and overall) by gender over a four-year period in the Australian labour market using the HILDA panel dataset. This paper contributes to the literature by building on initial Australian findings by Long (2005), which used HILDA’s Wave 1 dataset and concentrated only on overall job satisfaction.
We find females to be more satisfied with five of the six job satisfaction measures. Running gender separated random effects ordered probit models, we report that gender differences with different aspects of job satisfaction can be partially explained by both personal and labour market characteristics. In particular, job satisfaction for females is far less influenced by past labour market participation compared to males. Differences in workplace characteristics are less pronounced though unionised females are less satisfied at work compared to non-unionised females; a finding far less pronounced for males.
Given that past studies have found gender job satisfaction differences disappear for the younger and highly educated, we re-ran random effects ordered probit models for both ‘educated’ and ‘young groups’. Unlike previous studies, we find that younger females are still more satisfied at work compared to males in four of the six measures investigated. However, higher educated females are a much-different subset of employed females as a whole. They are only more satisfied with higher educated males with respect to pay, and are actually less satisfied with hours worked and job flexibility.
Overall, the use of HILDA’s panel dataset has produced results both consistent and inconsistent with previous findings in the literature. Females’ higher level of job satisfaction is found, with no econometric (but statistical) evidence that it is being fully eroded over time. However, we find no evidence that younger females exhibit job satisfaction rates comparable to young males, while highly educated females have satisfaction rates that failed to match initial expectations and deserve further investigation.
Personality Traits in HILDA
Ibolya Losoncz
For the first time in Release 5.0 of HILDA respondents were asked questions about their personality. Items were based on the Five Factor (Big 5) Personality Inventory, a descriptive model of personality identifying five broad dimensions (Extroversion, Agreeableness, Conscientiousness, Openness to experience and Emotional stability). The 36 items were single adjective items (such as talkative, orderly, kind) and were predominantly based on the 40-item Trait Descriptive Adjectives (TDA) developed by Saucier in 1994. Respondents were asked to identify on a 7-point scale how well each word described them.
This paper initially looks at the psychometric properties of the items. Factor analysis results will be reported and discussion will focus on how well the instrument supports the five personality dimensions factor structure it was designed to measure. Reliability of the subscales will also be reported and discussed.
The second part of the paper will examine how well the obtained personality subscales correlate with other measures (available in HILDA Wave 5) known to be associated with the Big Five Personality dimensions. The objective of this analysis is to evaluate the degree to which the five constructs in HILDA assesses the same constructs as those assessed by longer, well established and validated measures, thus providing some validity information for the scale.
The effects of family cohesion and personality on the mental health of young Australians
Michelle Tan
How can we explain the variation in mental health among youth? There is extensive research on youth outcomes and a large body of this work use family structure or divorce to proxy for instability (conflict) in order to explain why young people from broken homes have poorer life outcomes than other comparable youth. This paper investigates the determinants of mental health and departs from existing empirical studies in two ways. Firstly we derive a contemporaneous conflict index for the home based on the quality of the relationships among household members, rather than a simple indicator of martial status. Secondly, we include the ‘Big 5’ personality measures to control for those fixed factors that exist for youth from high conflict families and would give rise to selection bias. The ‘Big 5’ is assumed to remain stable over the time of the study. A sample of Australian youth between the ages of 15 and 25 is taken from the 5 waves of HILDA. By using various specifications we expect to find that interactions between levels of family conflict and personality traits help explain the differences in mental health.
Job Mobility in Australia
Nick Carroll and Jennifer Poehl
How important are different worker flows and what are the characteristics of those people who engage in job separations in Australia? This paper generates statistics on the level of job to job flows relative to other worker flows in Australia. The paper then investigates which groups are more likely to separate from their employers and how this varies depending on whether the move is voluntary or involuntary. This paper uses the first five waves of the Household, Income and Labour Dynamics in Australia (HILDA) survey. The key findings of the paper are that job to job flows are large and that the negative relationship between tenure and job separation appears to be the result of both person specific factors as well as match quality.
The effects of motherhood and job transitions on female earnings in Australia
Amanda Hosking
Utilising recent Australian panel data, this paper investigates how job transitions undertaken by Australian mothers around the birth of a child affect hourly earnings. I examine transitions between full-time and part-time hours and movement between employers. Both the economic theory of compensating amenities and institutional theories of labour market segmentation suggest that mothers who move to part-time hours or change employers after a birth will either experience a drop in hourly earnings or attract lower wage growth than mothers who return to work full-time and/or return to the same employer.
The analysis draws on data from the Household, Income and Labour Dynamics in Australia survey (HILDA), narrowing the sample to females who experience a birth over the course of the five waves of HILDA and who return to paid employment within three years. Changes in mothers’ hourly earnings are modelled using first-differences regression.
Results reveal that transitions from full-time to part-time working hours after a birth has a positive effect on wages. After adjusting for work experience and casual pay loadings, mothers who change from full-time to part-time work hours on average attract a 13 per cent increase in their hourly earnings. Mothers who return to a different employer after the birth neither experience an earnings penalty or premium. Supplementary analysis tests whether the effects of these two job transitions are similar for female employees at other stages across the lifecourse. These models reveal that childless women experience a rise in hourly earnings when moving from full-time to part-time hours that is on par with the increase employed mothers attract when making this same transition after a birth.
In summary, these findings run contrary to predictions that the transition to part-time employment by Australian mothers following a birth entails immediate downward mobility. These results differ from reported findings in Britain and the United States, and are likely to reflect differences in the nature and regulation of part-time work in Australia. However, there remains the possibility that mothers who interrupt employment to be a full-time carer for more than three years are disadvantaged on their return to employment in a part-time job or with a new employer. As additional waves of HILDA become available this analysis could be extended to include mothers with longer career interruptions and to explore the longer term consequences of part-time work on wage growth.
Labour Force Participation and Household Debt
Rochelle Belkar, Lynne Cockerell and Rebecca Edwards
In the past decade or so there has been a substantial rise in debt holdings and debt-servicing obligations of Australian households. This has been accompanied by a trend increase in labour force participation (LFP) for women and more recently for men. Microeconomic data shows a clear positive correlation between holding debt and LFP. This paper models the labour force participation decision of prime-age Australian women and men accounting for the influence of debt and assets. The potential two-way causation between debt and labour supply is also addressed. Data from the Household, Income and Labour Dynamics in Australia (HILDA) survey are used as the survey contains recent and detailed data on household wealth along with extensive labour market and demographic data. A cross-section model is estimated using the detailed measures of household debts and assets available in Wave 2 of the survey. In addition, a panel model, using only measures of owner-occupied housing debt and assets, is estimated using all five currently available waves.
Evidence is presented to suggest that increased indebtedness induces greater participation while the reverse effect, that greater participation gives access to credit, is not found to be important. In general, most of the effect of housing debt on an individual’s probability of participation in the labour force is captured through the household debt-servicing ratio, although the level of owner-occupied mortgage debt appears important for men. Also, the panel results suggest that accounting for unobserved heterogeneity across individuals is important when examining the influence of debt on labour supply.
Australian labour supply elasticity estimates: sensitivity to choice of model and sample selection across five waves of panel data
Robert Breunig and Joseph Mercante
Most reviews of labour supply elasticities acknowledge the wide range of wage elasticity estimates, both between and within various population groups. Also, most reviews acknowledge that the differences in estimates are predominantly due to differences in the way samples are selected, the time period chosen, and the modelling methodologies and techniques employed in each study. However, only a few studies investigate in greater detail the sources of these differences. In this paper we report on work currently being done using the Household, Income and Labour Dynamics in Australia (HILDA) Survey to explore the sensitivity of elasticity estimates to three different modelling approaches: a continuous hours model based on the Tobit approach; a continuous hours model based on a two-step sample selection approach; and a discrete hours model. These are estimated for five population groups: single males and females, married males and females and lone parents. Also, we evaluate the consequences of the treatment of non-workers, in particular the unemployed and the not in the labour force, in each of the models. We test whether differences in modelling choices about the way these two groups are treated have an influence on the elasticity estimates of each model. We address these questions using the first five waves of HILDA. This paper reports on the results to date using the two continuous hours modelling approaches.
A longitudinal analysis of income-related health inequalities in Australia
Jason D. Brandrup and Michael A. Kortt
The traditional approach to measuring income-related health inequalities has predominately relied on cross-sectional data to estimate measures of inequality such as concentration indices. Jones and Lopez (2003), however, have demonstrated that when there are differences in the health between those individuals who are upwardly or downwardly income mobile, longitudinal income-related inequality measures can show a greater degree of inequality compared to measures calculated using cross-sectional data.
Subsequently, Jones and Lopez (2003) developed a health-related mobility index to measure the difference between cross-sectional and longitudinal income-related health inequality concentration indices. Further, the authors decomposed their mobility index through simple linear regression into the contributions of different regressors. This approach was applied by the authors to nine waves of data in the British Household Panel Survey.
We replicated the method used by Jones and Lopez (2003). Specifically, we developed a health-related mobility index for Australia using self-reported physical and mental data (derived from the SF-36) from the first five waves (2001-2005) of the Household, Income and Labour Dynamics in Australia (HILDA) Survey. We also decomposed our mobility index to examine the contribution of different regressors.
Impacts of Individual/Household Socio-Economic Conditions and Area Environment on Health in Australia, Using Five-Wave Panel Data of HILDA Surveys
Ya-Seng (Arthur) Hsueh
Purpose and significance: This study aims to understand the relative importance of individual’s socio-economic conditions versus the living environment in a specific region to explain the differences in health status among people living in different geographic areas.
Past research has found a positive correlation between individual’s socio-economic status and health status as well as also regional variations in people’s health status. Then, in a longer term, between these two factors that are on different levels, which factor is more important in determining people’s health, i.e., individual’s social-economic factors or the living environment in a specific region? The answer for this question is not yet clear, because past studies were mainly based on cross-sectional data and the long term effects of individual socio-economic factors and regional environmental factors could not have been separated. Thus, there is an urgency and significance for this study, because it can help policy makers to decide the priorities in allocating limited public resources between improving individual socio-economic conditions versus health related infrastructures in specific regions.
Data and methods: This study will use five waves of yearly panel data of HILDA to study the above research questions with three justifications.
Firstly, the five-wave panel data can examine the longer term effect of individual’s social-economic status and the regional environmental factor on individual’s health status among those individuals who had not changed their working and living places between 2001 to 2005. Secondly, given that HILDA is a national representative data, I can use the Statistical Local Area (SLA) or Statistical Region (SR) defined in Australian Standard Geographical Classification (ASGC) to investigate regional variations of people’s health status across the five waves of surveys.
Secondly, having both individual social-economic and regional factors in one equation enables the interaction effect of these two factors on individual’s health to be isolated.
Finally, in order to account for possible cross-section heteroskedasticity and time-wise autocorrelation when conducting econometric analysis with panel data, panel-corrected standard errors will be used in analysis by employing Shazam econometrics software version 9. Since individual health and socio-economic factors may reinforce each other, seemingly unrelated regression (SUR) will also be used in the analysis.
Longitudinal Evidence on the Impact of Incarceration on Labour Market Outcomes and General Well-Being
Malathi Velamuri and Steven Stillman
The paper examines the impact of being incarcerated on labour market outcomes and general well-being using longitudinal date from the nationally representative Household Income and Labour Dynamics of Australia survey. We estimate OLS regressions on the entire population, propensity score matching models that examine outcomes for individuals that have been incarcerated compared to outcomes for individuals with similar observable characteristics over the same time period, and fixed effects regression models that examine changes in outcomes for individuals before/after incarceration relative to changes in outcomes over time for non-incarcerated individuals. The second and third types of models allow us to control for characteristics that may simultaneously cause certain individuals to commit crimes and put them at higher risk of poor outcomes. Our results indicate that incarceration has a large negative short-run impact on the likelihood that individuals are employed, but has a positive impact on total income. We also find that being incarcerated has a significant negative impact on mental health and life satisfaction, but leads to an increased satisfaction with family relationships.
Consequence or cause? Social disadvantage and teenage motherhood
Ann Evans
In 1968 Campbell stated that in the US:
The girl who has an illegitimate child at the age of 16 suddenly has 90% of her life’s script written for her. She will probably drop out of school…, not be able to find a steady job…; she may feel impelled to marry someone she might not have otherwise chosen. Her life choices are few, and most of them are bad (1968: 238).
These negative consequences reported by Campbell continue to be found in many cross-sectional studies in the US, Europe and the UK (for discussion see Berthoud and Robson 2001; SEU 1999; Luker 1996; Furstenberg 1987).
However, with advances in statistical modelling methods and the availability of longitudinal data, often of large scale and at the national level, researchers in the US and UK have been able to control for the difference in young mothers’ backgrounds. All of these studies have led to significant improvement in the methodology for studying teenage motherhood and pave the way for continued efforts in this area. On balance their results suggest that there are negative consequences associated with teenage motherhood, but these are much smaller than previously thought and can be moderated through social and personal support. Once we control for the selection effects of teenage motherhood it is suggested that teenage motherhood is a consequence, not just a cause, of social disadvantage.
In Australia Bradbury (2006) has attempted to control for the selection effect of motherhood on subsequent outcomes by using women who miscarriage as a comparison group. His findings suggest that teenage motherhood has no negative impact on education, labour market activity, income or location. In order to explore this issue further this paper uses a propensity score matching (PSM) technique on data from HILDA waves 1 through 5. This technique allows matching of women who have a teenage birth with those who do not based on variables observed prior to childbearing, thus separating out the effect of motherhood on subsequent life events such as training, employment, relationship formation and fertility. Using a PSM technique
Financial Poverty in Australia: Developing a Valid Measure Based on Wealth, Income and Consumption for use by Commonwealth and State Governments
Bruce Headey
National measures of poverty in Australia and other Western countries are usually based only on low income. But this is conceptually incorrect; the measures lack validity. To be poor is to have a low material standard of living – involuntarily. So measures of poverty should also take account of household consumption and wealth. If a household has an adequate level of consumption or a reasonable amount of wealth (net worth), it should not be classified as poor, even if it’s current income is low.
The invalidity of income-based measures has long been recognised in principle (e.g. Townsend, 1979; Ringen, 1987, 1988; Sen, 1999; Nolan et al, 2000; Saunders and Adelman, 2004; Saunders, 2005). In practical terms, the problem has been to combine measures of wealth, income and consumption in the same survey. The usual view, especially among Government statisticians, is that in order to measure consumption and wealth long batteries of questions are needed. Further, the standard view is that household expenditures/consumption can only be reliably measured using the ‘shopping diary’ method, which requires survey respondents to fill in a diary for a week.
The HILDA Survey team is attempting to overcome these practical problems. In the 2005 survey a battery of household expenditure items was included which, benchmarked against the ABS Household Expenditure Survey for 2003-04, appeared to provide accurate measurement of 57% of total household expenditure. These accurately measured items correlated 0.76 with total expenditure. On this basis, it is reasonable to estimate total household consumption from the HILDA data, and then to use the consumption measure in an improved measure of financial poverty.
In 2006 HILDA will, for the first time, provide a full set of household financial accounts. That is, household wealth, income and consumption will all be measured in the same survey. Wealth was previously measured in 2002 and the results matched well with aggregate measures in the National Accounts and with simulations developed by NATSEM (Headey, Marks and Wooden, 2005). Assuming that the 2006 measures also prove valid, it will be sensible to include them in a revised measure of financial poverty.
An important normative point is that poverty matters much more in the medium and long term than the short term. From a public policy and humanitarian point of view, the chief concern should be for people who remain poor for a substantial period. But standard cross-sectional surveys only measure poverty at one moment in time and so give a potentially misleading view. Panel surveys in many countries, including HILDA in Australia, show that most income-based poverty is short term (Headey, Warren and Harding, 2006). If HILDA proves able to provide a valid longitudinal series, based on defining poverty in terms of wealth and consumption as well as income, it will then be possible to assess whether poverty measured in a more valid way still appears to be mainly short term, or whether an improved measure leads to the conclusion that more people are medium or long term poor.
Clearly, much development work remains to be done, but the aim is for HILDA to produce valid medium and long term estimates of poverty, as well as short term. If this can be achieved, there would be a strong case for Australian Governments to make use of the revised poverty measures in evaluating tax, transfer and social policy measures.
The conference paper will:
• Present a case for preferring the revised approach to poverty measurement in preference to the four main available alternative approaches which have attracted serious attention from Western Governments. These are based on: (1) income (by far the most common approach); (2) capabilities (Sen, 1999); (3) social exclusion (European Commission, 2000) and (4) consumption deprivation indicators (Townsend, 1979; Saunders and Adelman, 2004; Saunders, 2005).
• Report construct validity tests showing that the revised measures correlate more highly with variables which would be expected to relate to poverty – including human capital variables and measures of financial stress – than income-based measures.
• Provide revised empirical estimates of financial poverty based on defining individuals as poor if they live in households with low wealth, low income and low consumption.
• Show that the short term (cross-sectional) poverty rate is much lower, using these measures, than if standard income-based measures are used. However, it is an open question whether medium and long term rates are lower or higher.
• Undertake analyses showing which groups are income poor but not consumption poor, and vice-versa. Wealth/asset poverty will also be analysed (Caner and Wolff, 2004).
• It may also be possible to provide the first preliminary longitudinal results, showing one and two-year poverty rates, using the improved measure.
Welfare Traps in Australia: Do they Bite?
Alfred M. Dockery, Rachel Ong and Gavin Wood
Minimising disincentives to employment participation that arise as a consequence of providing welfare has been a major objective of the recent welfare reform agenda in Australia. Of particular concern have been the high effective marginal tax rates (EMTRs) and replacement rates (RRs) facing welfare recipients as a result of the interaction of the tax and benefits system. Based on the financial-year income data contained in waves 1 to 4 of the Household, Income and Labour Dynamics in Australia Survey, the first part of this paper presents estimates of EMTRs facing all working-age Australians and of an alternative ‘disincentive’ measure, participation tax rates (PTRs), for non-working Australians. Significant innovations in the calculation of these measures are the incorporation of the effect of the withdrawal of rental subsidies for public housing tenants, which have not been included in the previous literature, and the allowance for changes in tax liabilities and benefit entitlements for the income unit rather than just for the individual. We find EMTRs and PTRs to have changed very little between 2000-01 and 2003-04, and to be particularly high for recipients of NewStart Allowance. In the second part of the paper we estimate the impact of EMTRs, the PTR and the more commonly used RR on the probability of transitions from non-employment to employment one year later. The PTR and RR are found to be superior to the EMTR in measuring disincentive effects for non-working persons. Second, these effects do matter. High PTRs or RRs have a moderate but significant effect on the probability that women will enter employment from outside the labour force, and a very large (negative) effect on the probability that an unemployed person will find work.
“I just want to get married - I don’t care to who!”: Marriage, Life Satisfaction and Educational Differences in Australia couples.
Shane Worner
Much of the empirical literature involved with marriage and partnership suggests that those individuals who are partnered exhibit increased life satisfaction compared to those who remain single. Coupled with this, we also observe that traits brought to a partnership exhibit non-random matching patterns, known as assortative mating. Following the existing literature we use the first 4 waves of the Housing, Income and Labour Dynamics in Australia (HILDA) Survey and standard panel data techniques to firstly investigate whether those individuals who are legally married exhibit significantly higher levels of life satisfaction. We then depart from the existing literature to investigate what type of partnership makes an individual happier and whether there are any indirect effects of spousal education. Results indicate that there are marked differences in the effects of marriage; spousal education and education difference on self reported life satisfaction. On average, married people are happier than their non-partnered contemporaries with average happiness increasing, and then decreasing, in the years preceding and succeeding the year of marriage. Of those who are married, the wife is more likely to report higher levels of happiness than the husband. In addition, the husband tends to be more educated than the wife, but the higher the education level of the wife, the happier the husband is. Further, the educational differences between the husband and wife seem to have no effect on happiness levels for both the husband and the wife.
Life In and After Being In A Small Business In Australia: Evidence From The HILDA Survey
Justin Craig, Michael Schaper, Clay Dibrell
Very little of the HILDA data set has been used to analyze the activities of Australian small business owner-operators, even though there are some 1.8 million small firms in existence. The purpose of this paper is two-fold. First, using four waves of the HILDA survey we investigate life in a small business in Australia. Specifically, we compare the reported life satisfaction and health of small business respondents with those respondents who are employees of private sector firms and government agencies. Second, we establish the effect that business closure has on family life, self-reported life satisfaction, and health. We establish from HILDA Wave 1 that 97 respondents reported that their small businesses closed. By tracking these respondents’ reported life satisfaction in the year of closure and in subsequent years, we are able to quantify the effect that business closure has on family life, individual life satisfaction, and health. We bifurcate the sample based on post-business closure respondent outcomes and link these results to theoretical discussions in the extant grief literature that suggest that the grief recovery process takes on either a loss orientation or a restoration orientation. We hypothesize that those who report stable levels of life satisfaction subsequent to the closure of their business are able to find suitable employment or were active in the pursuit of employment, whilst those business owners whose life satisfaction significantly declined as a consequence of business closure are either not active in pursuing employment or are not able to secure suitable employment.
The Impact of Housing Assistance on the Employment and Earnings Outcomes of Labour Market Program Participants
Simon Feeny, Rachel Ong, Heath Spong and Gavin Wood
This paper examines the employment and earnings outcomes of Australians undertaking Mutual Obligation Activities (MOAs). Specifically, it considers whether the outcomes for MOA participants vary according to whether they receive housing assistance. The research is motivated by evidence from the US which identifies superior employment and earnings outcomes of welfare recipients living in public or assisted housing. The paper utilizes all four waves of data from the Household Income and Labour Dynamics in Australia (HILDA) database. In Australia, housing assistance can be received in one of two specific forms: public housing; or Commonwealth Rent Assistance (CRA). Either of these forms of assistance might provide a disincentive for employment as individuals may lose their eligibility for assistance once their income reaches a certain threshold. On the other hand, the greater housing stability that can result for recipients of housing assistance may also assist in improving employment and earnings outcomes. For example, housing assistance can provide unemployed individuals the stability to focus solely on obtaining employment and taking advantage of that employment opportunity, without the additional concern of unstable housing tenure. A variety of models are posited which are estimated using different statistical techniques. Preliminary results suggest that housing assistance, in the aggregate and in its specific forms, has little impact (either negative or positive) on the outcomes of MOA participants. Further, there is evidence that outcomes for MOA participants undertaking employment and community participation programs and training programs are inferior to those MOA participants undertaking employment assistance programs.
Job Mobility and Segmentation in Australian City Labour Markets
Anthea Bill. William Mitchell and Riccardo Welters
In the last 15 years cities have developed a unique potential for achieving successful outcomes, by virtue of their scale, networks and advanced service functions. These attributes are said to have afforded city workers higher earnings and greater opportunity to appropriate productivity gains through job mobility.
However the benefits of job mobility arguably accrue only to those individuals located in dynamic local labour markets and in growing occupations with 'deep' skill-sets. The flip-side of flexibility is more insecurity, associated with casualisation and intense job-competition for low-skilled positions. Thus we might argue in cities, increased flexibility spurs career advancement in the primary segment of the labour market, while increased insecurity obstructs career advancement in the secondary segment. Work to date has identified that there are significant differences between cities and their non-metropolitan counterparts in terms of the motivations for job search and the nature of job transition, holding other factors.
This paper attempts to extend these findings using four waves of Household Income and Labour Dynamics Australia (HILDA) data, to examine whether cities do promote greater levels of mobility and whether primary and secondary labour market participants display different patterns of search and occupational transition in urban versus non-urban areas. Using a cross-sectional pooled dataset of the working age population from the four waves of HILDA (first in 2001 and last in 2004) we find clear evidence that job mobility is higher in metropolitan areas. Both an increased confidence that search will locate a new job and heightened fear of losing one's current job explain the higher levels of job mobility. As expected, confidence, linked to upward job mobility, is higher in the primary segment of urban labour markets. However surprisingly, fear of losing one's job, associated with downward job mobility, is also higher in the primary segment of the metropolitan labour market. This suggests that in urban areas the primary labour market may not be immune, and may in fact be particularly susceptible, to the adverse dynamics associated with increased job mobility. The exact reason for the greater *fear* of job loss amongst workers in the primary segment of cities requires further examination.
Achieved fertility, desires, expectations and intended number of children: a longitudinal analysis
Habte Tesfaghiorghis
The paper undertakes a longitudinal analysis of fertility using the first five waves of HILDA datasets. A balanced panel data of women aged 18-44 years was created for the purpose of the analysis. The paper has two parts. The descriptive part analyses changes in individual women’s marital status, number of children ever borne, desire to have children, expectations of having children and intended number of children between 2001 and 2005. This analysis will answer whether or not fertility desires, expectations and intended number of children were consistent over-time.
The analytical part fits longitudinal models to the number of children born to women, using appropriate models, to quantify the components of fertility change and the factors affecting fertility change.
This research is integral to the author’s previous research works on women’s fertility, and men and women’s fertility differences in achieved fertility, expectations and intentions, which were based on the cross-sectional 2001 HILDA dataset. Though the period covered is short, this paper will make the first attempt to our understanding of the dynamics of Australia’s fertility.
Contraceptive practice and the reproductive lifecourse
Edith Gray and Peter McDonald
This paper examines the contraceptive use of women in reproductive ages (18–44) using questions collected as part of the 2005 HILDA (Wave 5, Special Topic on Fertility). Previous research on contraceptive practice in Australia using representative surveys typically applies age as a proxy for reproductive history. We compare the patterns of contraceptive use found in HILDA to previous Australian representative surveys and find similar patterns of contraceptive use. Further, past research on contraceptive practice is extended by examining two key factors associated with contraceptive use: parity and fertility intentions.
Main findings indicate that the most commonly used methods are the oral contraceptive pill (25%), condom (19%) and methods of sterilisation (around 8% for female methods: tubal ligation and hysterectomy, and a further 9% for male methods: vasectomy of partner). The use of these methods varies substantially by the stage of a woman’s reproductive lifecourse: women use more effective methods (such as the contraceptive pill and sterilisation methods) when they have completed their family. Logistic regression models will be used to investigate the use of the three main contraceptive methods by parity and fertility intentions controlling for socio-demographic factors. We also investigate how method practice varies by when women state they want another child. This paper argues that while age and other socio-demographic factors are useful for understanding contraceptive use, investigations of contraceptive practice should also consider parity and fertility intentions.
Is Casual Employment a Bridge or a Trap?
Cameron Allan,Jenny Chalmers, Jeff Waddoups
Many advanced industrialized nations are experiencing an increase in nonstandard, precarious employment, which consists of part-time, contingent, and casual relationships between workers and their employers. It is thought that such relationships may be useful to employers because of the flexibility they provide in responding to increased competitive pressures. A potential cost of such precarious employment relationships, however, may be that they trap workers in substandard employment by reducing their attachment to work, limiting their access to training, and reducing their productivity and wages over the longer term. The proposed research will examine the conditions under which casual employment acts as a bridge to more standard full-time work, or alternatively, the conditions under which it reduces the likelihood that a transition to standard employment will be made.
The proportional hazard models we construct will be used to examine the nature and determinants of the transitions from casual employment to permanent employment. Hazard models can be estimated in a multivariate setting to relate the probability of exit from casual employment with time in casual work and individual and job-related characteristics. In addition we plan to estimate separate hazard models for various demographic groups to determine the extent to which casual employment affects outcomes differently across various groups.
The first four waves of the Household Income and Labour Dynamics in Australia (HILDA) will be used to estimate the proposed hazard models. The longitudinal nature of the data will allow us to determine the length of spells in casual employment, which can then be linked to characteristics thought to affect labor market transitions, such as age, experience, tenure with current employer, gender, union status, etc. The results will increase our knowledge about the potential impacts of casual employment on labor market outcomes and careers paths of workers in Australia, and thus on the possible costs associated with movement toward nonstandard, precarious employment relationships. As such, the research has the potential to inform policy-makers about whether there is a need to revisit policies aimed at regulating the use of casual employment.
Transitions from Casual Employment in Australia
Hielke Buddelmeyer
Strong claims are often made about the harmful effects that casual employment can have on future employment prospects, yet serious research on this issue in Australia has been relatively scant. This paper seeks to help redress this deficiency. Specifically, it uses longitudinal data from the Household, Income and Labour Dynamics in Australia (HILDA) Survey to examine the rate of mobility in and out of casual employment, and where casual employees go when they exit casual employment. Personal characteristics associated with mobility in and out of casual employment are identified. By exploiting the longitudinal aspect of HILDA, inclusion of random effects help estimate the extent to which casual employment, and transitions out of casual employment, are the result of employee preferences for different types of employment.
The Rise of Three- and Skipped-Generation Australian Households
Peter Brandon and Lyn Griffen
Considerable increases in the numbers of grandchildren living with grandparents have prompted concerns over their economic wellbeing and grandparents’ access to additional resources. Using five waves of data from HILDA, we examine the rise of three- and skipped generation households among Australian children and analyze the factors that accelerate the formation of such living arrangements. We expect our findings that are based upon survival models to suggest that family structure is pivotal to understanding differences in the rise of three- and skipped-generation households as well as the prior economic disadvantage among households. We also expect to find that grandchildren who eventually live with grandparents to be the most likely to be poor.
Joiners, leavers, stayers and abstainers: Private Health Insurance choices in Australia
Stephanie Knox, Denzil Fiebig, Vineta Salale, Elizabeth Savage
The percentage of Australians taking up Private Health Insurance (PHI) was in decline following the introduction of Medicare in 1984 (PHIAC). To arrest this decline the Australian Government introduced a suite of policies, between 1997 and 2000, to create incentives for Australians to purchase private health insurance. These policies include an increased Medicare levy for those without PHI on high incomes, introduced in 1997, a 30% rebate for private hospital cover (introduced 1998), and the Lifetime Health Cover (LHC) policy where PHI premiums are set at age of entry, increasing for each year older than 30 years (introduced 2000).
In 2004 (Wave 4) the longitudinal study on Household Income and Labour Dynamics in Australia (HILDA), included a series of questions on private health insurance and hospital use. We used the HILDA study panel to investigate the demographic, health and income factors related to the PHI decisions, especially around the introduction of the Lifetime Health Cover policy. Specifically we were interested in who was most influenced to purchase PHI in 2000 as a response to the Lifetime Health Cover policy deadline. Are those who have joined PHI since the introduction of LHC different from those who joined prior to LHC. What are the characteristics of those who have dropped PHI since the introduction of LHC?
The outcome was self-report of PHI choices at interview in 2004. Predictors were responses in Wave 1 (2001) and Wave 2 (2002) interviews. We used the balanced panel of respondents who had complete data for the relevant variables in Wave 1 to Wave 4 (n = 9,204). We initially modelled the PHI outcome using multinomial logit and then refitted the model to allow for heterogeneity of choice and correlation across alternatives. After controlling for other factors, we found that LHC prompted well-off working age adults (30-55 yrs) to purchase before the 2000 deadline. Young singles or couples with no children, and the overseas born were more likely to purchase since 2001. Rapidly increasing income and prosperity is an important factor in the decision to purchase PHI since 2001, while the relatively less well-off still drop PHI in spite of current policy incentives.
The Survey of Family, Income and Employment (SoFIE)
The Survey of Family, Income and Employment (SoFIE) is the first official national longitudinal survey on individual and family dynamics in New Zealand. The survey consists of eight waves of an annual cycle, from 1 October 2002 to 30 September 2010. Its aim is to look at changes in income, labour force participation in association with changes in individual, family and household circumstances.
The presentation provides an overview of SoFIE, including its inception, survey design and processes. This is followed by an assessment of the current activities and outputs. The presentation concludes with a view on what is planned in the short to medium term.
The New Zealand Linked Employer-Employee Data (LEED)
The Linked Employer-Employee Data (LEED) project integrates longitudinal data on payments made to employees from which tax is deducted (sourced from Inland Revenue) with Statistics NZ’s longitudinal business frame data. LEED allows us to produce new and enhanced statistics on labour market dynamics, sources of income, income transitions, income spells and multiple job holders.
The presentation provides a brief history of LEED, an overview of published statistics and some findings. This is followed by a summary of the current research programme and benefits of LEED. The presentation concludes with a summary of planned enhancements to LEED in the medium to long term.
|