Social Statistics Seminar Series - Lorraine Dearden
| Dates: | 12 November 2024 |
|---|
| Times: | 14:00 - 15:30 |
|---|
| What is it: | Seminar |
|---|
| Organiser: | School of Social Sciences |
|---|
| Who is it for: | University staff, External researchers, Adults, Alumni, Current University students, General public |
|---|
| Speaker: | Lorraine Dearden |
|---|
|
|
The impact of violence on partnered women’s labour market outcomes: evidence from Australia
Longitudinal survey data measuring domestic violence (DV) is very rare which makes looking at the short- and medium-term impact of the economic implications of DV in a reliable way difficult. One exception is the paper by Adams-Prassl, Huttunen, Nix and Zhang (2024) which uses unique administrative data from Finland. They find that women in abusive relationships suffer large and significant earnings and employment falls. In most countries, however, rich panel data on DV does not exist. This paper uses rich household panel survey data from Australia (HILDA data), to see if we can replicate these findings from Finland when questions on physical violence (PV) but not DV are asked. We compare the incidence of PV for partnered women who ever separate and those who never separate and see that 23.8% of the former group have experienced at least one episode of PV whereas in the never separated group it is 5.6%. From cross-sectional surveys in Australia (Australian Bureau of Statistics Personal Safety Survey (PSS)) it is estimated that around 1 in 4 women have experienced violence from an intimate partner since the age of 15 (PSS 2021-22). We speculate that most of the difference between these two groups is due to differences in the rates of DV among the two groups.
We use a matched event study staggered difference in difference design as proposed by Callaway and Sant’Anna (2021) for our two groups. Whilst we find no significant effects of PV for our never-separated group of partnered women, we find significant impacts of PV for our ever-separated group, not only in the short-term but over the whole 5-year period after the first incidence of PV. After the first incidence of PV whilst partnered compared to immediately before, such women are (using 5-year weighted averages):
(i) more likely to reduce full-time employment by around 5.5 percentage points (an 18% decrease);
(ii) more likely to stop participating in the labour market with the proportion not in the labour force increasing by 5.6 percentage points (a 16% increase);
(iii) more likely to reduce the hours they spend in the labour market by around 2.2 hours per week (a 12.4% decrease);
(iv) more likely to have lower income from wages and salaries of $3,356 per year (a 13.2% decrease), lower household gross equivalised income of $8,027 per year (a 15.0% decrease), lower household disposable equivalised income of $5,382 per year (a 11.9% decrease), and a very significant reduction in family adjusted taxable income of $27,610 per year (a 44.9% decrease); and are
(v) more likely to see an increase in Australian government benefit transfers which is the result of their eligibility for and amounts received due to lower means-tested incomes with an estimated increase of benefits of $4,305 per annum (a 34.2% increase).
Like Adams-Prassl et. al. (2024) we uncover compelling evidence of negative impacts of likely DV on both labour market participation as well as income and earnings for ever separated women. Critically, the effects of DV are not temporary and seem to imply negative economic effects for at least 5 years.
Speaker
Lorraine Dearden
Role: Professor of Economics and Social Statistics
Organisation: University College London
Travel and Contact Information
Find event
Arthur Lewis_2.016/2017 Boardroom
