Dr. Claire Keane:

I thank the Chairman for inviting us. Today, I will discuss our forthcoming report that carries out a gender impact assessment of the tax benefit policy in Ireland over the past ten years. The work has been funded by the Parliamentary Budget Office and the motivation behind the work is the commitment in the programme for Government to being able to gender proof budgets. We believe this is making a practical contribution in terms of being able to do that.

To give an overview of what I will talk about today, I will touch briefly on gender impact assessments and what the ESRI has done to carry one out. I will present some results. I know that we, in the ESRI, like to bring lots of numbers with us but I promise there will not be lots of numbers. We can happily forward the members a copy of the report, but I will give them an overview of the results from the report and finish with some conclusions.

What is a gender impact assessment? The European Commission would say it is the estimation of different effects of a policy, be they positive, negative or neutral, in terms of gender inequality. I am aware the committee has been looking into the area of gender budgeting, so we see this as being part of that gender budgeting process.

The aim of carrying out a gender impact assessment is to provide evidence to policymakers. It is to try to identify if there are potentially negative effects of policies and, in having that awareness, if something can be done to reduce those negative effects. It is to try to help and improve decision-making and planning.

Internationally, we know that approximately half of OECD countries are doing some form of gender budgeting.

Looking at work the committee has done in the past, it is clear that there is quite a variation in what exactly is being done. Our report uses the ESRI tax-benefit model, known as SWITCH, to carry out a gender impact assessment of tax-benefit policies. Some members may be aware of the model which has existed for a number of decades in Ireland. It takes a representative sample of the population and one can use it to determine the distributive impact of tax-benefit policy changes. To date, members will probably have seen it in use around budget time. It allows us to see what changes happened and who they impacted. Generally, we look at deciles and whether poorer or richer people were impacted by tax-benefit policy changes. Are different families affected differently? Is it couples with one or two earners or single persons? That is what it has been used for to date. The addition we have made to the model now allows us to look at the impact of policy according to gender.

The model covers income tax, the universal social charge, PRSI, cash benefits and some non-cash benefits also. There are two main advantages to having the model available and integrating gender impact assessment within it. First, the model is already in use in government in five Departments. As of this morning, a new version of the model was circulated which integrates gender impacts. In doing this work, we have allowed for it to continue to be done in future within Government. The second advantage of the model is that it allows one to look at changes before they are made. If one is trying to gender proof a budget, one can carry out an analysis before any changes are made to determine what the impact of tax-benefit policy changes will be by gender. That can help to inform one's decisions.

One might ask why tax-benefit policies would have any gender impact. It is not as if there are different tax rates or benefits which men and women receive, with the exception of maternity or paternity benefit. However, tax-benefit policy will have different impacts according to gender because men and women tend to occupy different roles in society. For example, there are fewer women in employment and more of them are involved in caring roles than are men. We saw in recent years in relation to pension coverage that because women may tend to spend more time out of the workplace and have fewer social insurance contributions, they may be entitled to a lower level or rate of benefits on foot of policy changes. All of these differences between men and women lead to different incomes and entitlements to benefits and, therefore, to differing impacts from tax-benefit policy changes. For example, 99% of one-parent family payment recipients are women and 75% of carer's allowance recipients are women. Men are more likely to receive jobseeker's payments, in particular long-term payments, and are also more likely to be in receipt of disability allowance. Those are some of the benefit differences we see, but we also see differences in income. Men are more likely to be in employment and to work full-time than women and where women are in the workforce, it is more likely to be in part-time employment. There are also gender pay gaps whereby women are twice as likely to receive the minimum wage. All of those differences in income are going to lead to differences in tax liabilities. With a progressive tax system, any changes on the tax side will have a stronger effect on men than on women. It is not entirely the case that we are looking at negative effects on women. There could be negative effects on both genders of different changes.

In carrying out this work, it is pretty straightforward to consider single men and women and to determine how tax-benefit policy impacts on them. Where they are single, we do not have to worry about the fact that they may be living with partners. As such, we focus on single persons and on couples separately. When we look at couples, it is much more difficult to determine what the gender impact of a tax-benefit policy will be. Where people live as a couple, it is difficult to say whether a policy impacts on both persons equally. For that reason, we looked at two extreme examples. We looked at what we call "full income sharing" where we assume couples share all of their income fully and then we look at the other extreme example whereby it is assumed that they do not share any income. Members can think about this in the context of their own lives. If one is living with one's partner, does one put all one's money into a joint bank account or keep some or all of it in a separate account? Does one party pay the mortgage while the other person pays other bills with things being evenly balanced as a result? We never know fully how couples share which is why we have modelled these two extreme examples.

A simple example in this context might involve a man earning €30,000 who pays income tax and USC. A policy change may change his income. For example, if the standard rate of tax is increased by 1%, it will result in a reduction of approximately 1% in his take-home pay. On the other hand, one might have a woman who is not working but is caring for an elderly relative and receiving carer's benefit of approximately €11,000 a year. If a policy change reduces carer's benefit by €5, it will result in a reduction in her income of approximately 7%.

If that is applied to a couple, that averages out at around 3% between them, if we assume that they share all those losses in the same way. That illustrates that changes can have more of an impact on one member of a couple than another.

What do we think is more realistic? Previous ESRI research has tried to look into this and see exactly how couples share income. We tend to find in the vast majority of cases that couples share a large proportion of their income - not all of it, but a good chunk of it. That is probably a more realistic option. Who gets what income can matter in terms of how it gets spent. If one partner is without an independent income, that might have an impact on that partner weighing in on decisions about how the money is spent in the household. If one person is not working and one person is working, maybe the person working has more of a say on how money gets spent.

There is research on this. In the UK, for example, they changed the child benefit payment. It used to be paid as a tax credit, so generally the man used to get it. It was changed to a benefit that the mother would generally receive. Looking at how money was spent when those changes happened, women were spending more money on children and also spending more money on themselves. Who gets what can matter and that is why we have done these two extreme examples.

In terms of our results, we have focused in the report on how tax benefit policy has had an impact on the disposable income of men and women over the past ten years. There is no real gender impact among single people. There was not much of a difference in the impact of tax benefit policy between men and women who were single. Singles with children lost out more so than singles without children. That was largely driven by cuts to child benefit and other welfare cuts. There was again no huge gender difference, but it is worth bearing in mind that single, lone parents tend mainly to be female rather than male.

Couples, assuming that they shared all their income, did not lose out much more sharply than singles. In terms of the individual income within couples, however, there were differences. In the case of couples without children who were not sharing their income at all, the impact on them of tax benefit policy was roughly equal. In the case of couples with children, there was a gender impact. Women were more affected. Their individual incomes were cut more sharply due to tax benefit policy. It is noticeable that those differences were driven by the lower part of the income distribution. If we take those cases where women were losing out more than men, it was lower income women losing out more so relative to men.

That illustrates those differences between men and women and the different roles they occupy. We looked at results by economic status to see whether there were differences if we compared working men with working women. There was not much of a difference there. The differences were being driven by women out of work compared with men out of work. If we know that neither men nor women in work were losing out more due to tax benefit policy, but we know there are many more men in work than women, those different roles and those different employment rates of women can account for a lot of where these gaps are coming from.

A positive can be taken from that. Female participation rates have been rising in recent years so that would suggest that any gap in the impact of tax benefit policy by gender might close as female participation rates rise.

I did not want to get too bogged down in the detail and results of this work because it is looking back over a time period. This tool is now available. It is a practical tool in how to implement gender budgeting and carry out gender impact assessments. The advantage is that it can be used in advance of changes happening and before policy has been decided. It is available to Departments so this could continue. For example, it is in use in the social impact assessment. The model is used to see if certain income groups have lost out by more or less due to policy. It could be an addition there.

The ESRI focused on gender proofing of budgets and gender impact assessments. The programme for Government discusses wider equality proofing and this method of separating people by gender could also be done on other grounds. It could look at details by age, family status and disability status. There is potential to expand it out even further.

If the committee has any questions, I would be happy to-----