Women's economic empowerment and domestic violence

The safety of women in India – both inside and outside homes – is a major concern. This column explores the link between women’s economic empowerment, in the form of stronger inheritance rights and working status, and the incidence of domestic violence. It suggests that empowering women through income and wealth reduces the likelihood of them becoming victims of domestic violence.

Crime against women is a particularly severe problem in India. A survey conducted by Thomson Reuters Foundation in 2012 ranks India as the worst country for women among the G-20 nations. Data from India’s National Crime Records Bureau (NCRB) suggests that there were 244,270 crimes committed against women during 2012 (a rate of 41 crimes per 100,000 women). These crimes include 24,923 rapes (4 per 100,000 women), 8,233 dowry-related murders (1 per 100,000 women), and 106,527 instances of abuse by a husband or his relatives (18 per 100,000 women). Further, as per the National Family Health Survey (NFHS), 36% of ever-married women have experienced some form of physical abuse from their spouse, such as being pushed, slapped, punched, kicked, strangled, burned, or threatened with a weapon. Moreover, almost three quarters of women who have experienced violence report that they have never sought help. This suggests that the NCRB database suffers from severe under-reporting and that the problem of violence against women is even more prevalent than what NCRB suggests.

Is there a link between women’s economic empowerment and domestic violence?

I, along with my co-author Sita Slavov, assess the link between economic empowerment of women through work, earnings or wealth, and the incidence of domestic violence (Mathur and Slavov 2013). On the one hand, higher levels of income and wealth of women may reduce domestic violence because they improve the bargaining position of women in the household. On the other hand, husbands may view more empowered wives as a threat to their own status, provoking more retaliatory violence.

To explore this issue, we analyse data on ever-married women from two household-level datasets - the NFHS and the Indian Human Development Survey (IHDS). We use variation in state-level inheritance laws as an exogenous proxy for women’s economic status. The inheritance law changes that we study relate to the Hindu Succession Act of 1956 – which applies to Hindus, Buddhists, Jains, and Sikhs – and which put daughters at a disadvantage relative to sons with respect to the inheritance of family property. Under the Act, daughters had the same right as sons to inherit their fathers’ separate property, as well as the father’s “notional” share of the family’s ancestral property. However, sons were further entitled to their own independent share of the family’s ancestral property. Sons were also allowed to request that the ancestral property be divided, while daughters had no such right.

After the 1956 Act, some states enacted legislation to amend the Act and make property inheritance laws more gender neutral. These states include Kerala (1976), Andhra Pradesh (1986), Tamil Nadu (1989), Maharashtra (1994), and Karnataka (1994). These amendments gave women an independent claim to the family’s ancestral property. However, the amendments only applied to women who were not married when the legislation was passed. These state-level changes in inheritance law provide a ‘natural experiment’ for our study, as they allow us to test whether women who were affected by the amendments (‘treatment’ group) in these states experienced different outcomes compared to women who were unaffected by the amendments (‘control’ group). The ‘control’ group therefore includes women who were married at the time this legislation was passed in the five states listed above as well as women in all other states that were still subject to the 1956 inheritance law. Changes to property inheritance law should be impactful because a large share of wealth in India consists of land (Roy 2008).

In addition to changes in inheritance laws, we test whether other measures of economic empowerment, such as having a job and having high earnings from the job, are associated with better outcomes for women.

Evidence from IHDS Data

IHDS is a nationally representative survey covering 41,554 households in 1,503 villages and 971 urban neighbourhoods across India. The survey covers the period 2004-2005 and provides an extensive set of data on issues relating to domestic violence and women’s earnings and employment. The IHDS asks eligible women whether, “in your community, it is usual for husbands to beat their wives” for a variety of possible reasons, including leaving home without notifying him, failing to pay a dowry, neglecting household responsibilities, not cooking meals that are up to standard, and having extramarital affairs. We consolidate the responses to all of these questions to construct our outcome variable, which is an indicator that takes the value one if the wife answers “yes” to any of these questions. The outcome variable takes the value zero if the answer to all of these questions is “no”.

Our results suggest that there is no statistically significant relationship between reported beatings and either work status or hours worked. However, women with higher earnings are marginally less likely to report that beatings are common. In particular, an increase of Rs. 10,000 (a large increase relative to the average woman’s earnings) reduces the probability of reporting that beatings are common by 0.5 percentage points. In comparison, an increase in family income other than the woman’s earnings is associated with a smaller (but still statistically significant) reduction in the probability of reporting beatings.

We find no evidence of a relationship between reported beatings and amendments in the inheritance law. One reason for this may be that the IHDS asks whether violence is common in the respondent’s community, rather than whether the respondent has had personal experience of violence. If a community includes women from both the treatment (affected by amendments in inheritance law) and control (not affected by amendments in inheritance law) groups, the treated women may report that they have heard that violence is common even if they have never been personally victimised.

While the IHDS results are consistent with the story that women’s empowerment – through employment and earnings – lowers the risk of domestic violence, we cannot establish that this is a causal effect. In particular, there are likely to be many unobservable factors (for example, the husband’s attitude towards women) that jointly determine both a woman’s earnings and her risk of being beaten.

Evidence from NFHS data

One limitation of the IHDS is that it has data from only one round of survey. Given that NFHS is undertaken every few years, we can also look at both within-state and over time variation to identify the effects on domestic violence. We work with data from the surveys conducted in 1998-99 and 2005-2006 since these contain information on domestic violence.

We use two alternative outcome variables in our NFHS analysis. The first is an indicator variable with the value one or zero based on whether the respondent has ever been subjected to violence by her spouse. The second indicator variable is whether the respondent has ever been subjected to violence by anyone (including her spouse). For factors that may influence domestic violence, we construct three measures of women’s empowerment: whether or not a woman is employed, type of work performed (seasonal, year-round etc.), and occupation in the previous year. Unfortunately, NFHS data does not provide information on the earnings of the responding women or their families, so the above three measures are used as proxies for earning. In addition, we analyse the effect of change in inheritance laws on domestic violence.

Our results suggest that women who were beneficiaries of the inheritance law are significantly less likely (2.5 percentage points) to be beaten by their spouses or by any other person compared to women who did not benefit from the law. These results are not substantially altered by including all three empowerment measures simultaneously. In addition, we find that women with more education, women whose families have a higher standard of living, and women who married after age 18 are less likely to be beaten. We obtained similar results for these factors in the IHDS data analysis as well.

More power to women

Our results from the IHDS data are supportive of the effect of incomes and jobs on reducing domestic violence against women, while the results from the NFHS data are more conclusive about the effect of changes in wealth, through inheritance laws, on reducing spousal violence.

There are a number of channels through which women’s inheritance rights and earnings might influence the probability of them experiencing domestic violence. Increase in wealth through inheritance rights and increase in income through earnings from a job, are likely to increase the bargaining power and importance of women within the household. Even in cases where actual inheritance has not occurred, the probability that a woman could inherit land and property may have a similar effect.

In other research focusing on inheritance law changes, Deininger et al. (2013) suggest that strengthening a woman’s inheritance rights may alter her marriage market outcomes. Their results suggest that the state-level amendments to the Hindu Succession Act tended to increase the age at which women would marry, particularly in families where the father was more educated and therefore more likely to be aware of legal changes. Later-marriage ages are associated with better marriage outcomes for women. For instance, Field and Ambrus (2008) and Jensen and Thornton (2003) have shown that a woman’s age at marriage affects her educational attainment, bargaining power, likelihood of experiencing domestic violence, and control over reproduction.

In our context, it is possible that a woman with stronger inheritance rights or higher incomes may be more likely to marry a non-abusive spouse. While the actual causal mechanism is unclear, overall, the data provide strong support for the hypothesis that empowering women through income and wealth reduces the likelihood that they will become a victim of domestic violence.