![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
Abstract
This article explores the impact of Self-Help Group (SHG) participation on the frequency of domestic violence in rural India. The study hypothesizes that SHG participation can raise tensions between married men and women because husbands may perceive some aspects of women’s empowerment as a challenge to patriarchal cultural norms. Using household panel data collected in rural Andhra Pradesh in 2004, 2006, and 2007, this article employs double difference methodology with an instrumental variables approach for impact evaluation. The estimation results show that, while SHG participation reduced domestic violence in the short-term, medium-term participation increased the frequency of domestic violence, particularly after women’s credit access through SHG participation had improved. This article furthermore reveals that the impact of SHG participation on domestic violence was more pronounced among couples who married with dowry. Spouses who practiced dowry appear to be more susceptible to financial inflow through the wife.
HIGHLIGHTS
Self-Help Group (SHG) participation impacts the frequency of domestic violence in conflicting ways.
Women’s SHG participation initially reduces tensions with their husbands.
In the medium term, women’s access to credit creates conflicts with their husbands.
SHG participation alone is not enough to overcome patriarchal practices and structures.
Effective gender-advocacy programs should include training to change both women’s and men’s attitudes.
Acknowledgments
We are grateful for valuable comments from Yana Rodgers on an earlier version of this paper presented at the IAFFE sessions at the 2018 ASSA Annual Meeting in Philadelphia. We also thank Ronni Alexander, Koji Yamazaki, Keijiro Otuska, and Saumik Paul for their insightful comments and suggestions. We wish to thank Renate Kloeppinger-Todd of the World Bank and the government of Norway, who provided the financial support for the research on which this article is based. We also wish to thank Shaik Galab, P. Prudhvikar Reddy, and Yanyan Liu for their support in data collection and field studies. We also acknowledge the helpful suggestions from the two anonymous reviewers. Needless to say, any remaining errors are the responsibility of the authors. All personal information that would allow the identification of any person(s) described in the article has been removed. The Institutional Review Board at the University of Wisconsin–Madison approved the research protocol in this paper on June 20, 2006. The protocol number is SE-2006-0354.
SUPPLEMENTAL DATA
Supplemental data for this article can be accessed at https://doi.org/10.1080/13545701.2021.1987499https://doi.org/10.1080/13545701.2021.1987499.
Notes
1 See Naila Kabeer (Citation2001) for a review of some of these conflicting findings.
2 Andhra Pradesh, Tamil Nadu, Uttar Pradesh, Orissa, and Maharashtra.
3 In June 2014, Telangana became independent from AP as the twenty-nineth state of India. When our surveys were undertaken, these two states comprised of one state, AP.
4 We reserve the use of the term “long-term” in this study for outcomes or impacts that are a result of stable changes within the socioeconomic context brought about by social movements and advocacy as well as policies and program interventions. Short-term and medium-term changes (or non-changes) refer to impacts that depend on resources and activities that may well be temporary.
5 Strategic choices are distinguished from practical choices by the fact that they affect intrinsic elements of gender inequality and challenge gendered power as compared to practical or instrumental choices that may improve their gendered conditions but do not question their status nor the status quo (Moser Citation1993).
6 Domestic violence is defined by the United Nations in its Declaration on the Elimination of Violence against Women as “Physical, sexual and psychological violence occurring in the family” (UN Citation1993). The World Health Organization’s numerous studies on domestic violence (intimate partner violence) define it as “violence against women by male intimate partners” including physical and sexual violence, emotional abuse, and controlling behaviors (WHO Citation2005: 5).
7 Most studies that consider the effect of SHG participation on domestic violence have only very short-term domestic violence data (based on one- to 1.5-year recall periods). This article utilizes short-term (2.5 years) and medium-term (five years) data on domestic violence.
8 For example, the Pre-Conception and Pre-Natal Diagnostic Techniques Act of 1994 bans prenatal sex determination in order to reduce female feticide; the Prohibition of Child Marriage Act of 2006 increased the age of marriage for women to 18 and for men to 21; the Hindu Succession Amendment Act of 2005 grants daughters equal property inheritance as sons; the Dowry Prohibition Act of 1961 prohibits and criminalizes the giving or receiving of dowry; and the Protection of Women from Domestic Violence Act of 2005 gives the court power to pass protection orders for women from their abusers.
9 K. M. Rabiul Karim and Chi Kong Law (Citation2016) used husband’s gender ideology to predict whether a wife controlled microcredit or turned it over to her husband to use.
10 Social organization included grouping SHGs at village and mandal levels as well as forming resource, producer, and labor associations within villages.
11 The environment in which MFIs developed, particularly those set up by for-profit institutions, encouraged clients to over-borrow resulting in high levels of debt stress and suicides, especially in peri-urban and urban areas. As a result, the AP state government enacted the Andhra Pradesh Microfinance Institutions Act (regulation of money lending) in 2010 (Bellman and Chang Citation2010).
12 These six districts are: Adilabad, Anantapur, Chittoor, Mahbuhnagar, Srikakulam, and Vizianagaram. The program was expanded to all twenty-two state districts under the Rural Poverty Reduction Programme (RPRP) beginning in 2003–04.
13 Some MFIs that utilize group collateral form smaller groups such as BASIX (4–6 persons), SHARE (5 persons), CDF (5 persons), and SKS (5 persons).
14 Other NGOs as well as banks and private financial institutions in AP have also formed SHGs with the objective of providing microfinance services.
15 One of them is life and health insurance program based in the community in order to improve preventive and ambulatory health care as well as hospitalization. Another program to improve food security is a rice credit line that provides high-quality rice at lower prices than those available at local stores.
16 SHG women can participate in training programs such as bookkeeping and develop entrepreneurial skills and capacity. There are also other less tangible benefits, for example, illiterate women develop communication skills in order to negotiate and learn to sign their names on contracts.
17 The panel monitoring and evaluation study was financed by the World Bank.
18 First, based on the information provided by the Society for Elimination of Rural Poverty (SERP), which was responsible for the implementation of DPIP in the field, households were stratified into four economic class categories by monthly per capita consumer expenditures: poorest of the poor, poor, not so poor, and not poor. Then, a total of ten households were randomly selected: four households from the poorest of the poor, three from the poor, two from the not so poor, and one from the not poor. A second survey was undertaken in mid-2006 to re-interview the 2004 survey households.
19 Klaus Deininger and Yanyan Liu (Citation2013a) by using the same data set showed that program-mandal households were more likely to have larger household sizes, more secure landownership, and a higher proportion of the very poor.
20 We consider mature SHGs those that has been active for more than two and a half years. In fact, there are several challenges in conducting impact evaluation of SHG participation with the data available to us. A major challenge is that the first round of the household survey for monitoring and evaluation of the DPIP program was conducted in 2004 after the program had begun to be introduced in the control mandals. As a result, some 2004 survey households in the control mandals had recently become a member of a SHG. Moreover, by the time of the second survey round in 2006, SHGs were present in nearly the entire state, including the control mandals, and other microfinance institutions were increasingly servicing rural areas. It was not possible, therefore, to have matching treatment and control areas even for our sub-sample. Nonetheless, most SHG participants in the control mandals had been members for less than one year, while the majority of SHG participants in the program mandals had been members for more than two years at the time of the first-round survey in 2004. Hence, it is still possible to compare the outcomes of medium-term participants in the treatment mandals to those of the shorter-term participants in the control mandals. For the reason mentioned above, this article tries to estimate the effects of participating in a SHG for more than two and half years, which is the same approach employed by Klaus Deininger and Yanyan Liu (Citation2013a).
21 The 5.9 percent of the households in the control mandals that used to be members of the pilot program (Development of Women and Children in Rural Areas, DWCRA) and thereafter converted into members of the DPIP program are considered to be participants in SHGs for more than two and half years in 2004.
22 During the corresponding periods, DPIP credit use levels among SHG members also increased. The percent of households accessing DPIP credit increased to 36 percent in 2004 and 48 percent in 2006 (see Online Appendix Table A1). Loan disbursements expanded very rapidly, and credit uptake in the treatment mandals was higher than that of the control mandals in 2004, whereas the credit uptake rate in the control mandals in 2007 was almost the same as that of the treatment mandals.
23 Scheduled castes are low-status communities within the framework of the Hindu caste system who have historically faced deprivation, oppression, discrimination, and extreme social isolation. Scheduled tribes are tribes that have suffered discrimination and disadvantage because of their ethnicity and are generally geographically isolated. Other backward castes are communities that have been historically marginalized and discriminated against and continue to experience social isolation but not as severely as scheduled castes. Other caste groups are Hindu communities that enjoy higher social status than those defined here.
24 At all the research stages, we followed the guidelines articulated in “Putting women first: Ethical and safety recommendations for research on domestic violence against women (2001)” and “Researching violence against women: a practical guide for researchers and activists (2005),” published by WHO. Our research proposal and research instruments were ethically assessed and approved by the Institutional Review Board in University of Wisconsin-Madison. For the interviews, women enumerators who belonged to similar social categories, that is caste groups, were recruited for women respondents, and enormous efforts were made to train them prior to the interviews. During all the survey processes, we tried to minimize the underreporting issue, yet because the enumerators did not force the respondents to give answers, the reporting rate of domestic violence could be lower than the reality. For the purpose of the impact estimation, the underreporting issue does not yield any bias with the Difference-in-differences (DID) analysis because the underreporting issue is highly likely to happen in the same or similar manner in both treatment and control mandals.
25 The DID estimate between 2004 and 2006 is obtained by a comparison between the program mandals where the DPIP had been present for approximately five years and the control mandals where the DPIP had been present for about two and half years. If the presence of DPIP for two and half years in the control mandals also reduced the frequency of domestic violence, our estimate is highly likely to be biased upward. Yet, it seems that such a reduction is unlikely in the control mandals because SHG activities were implemented at a faster rate and loan were also disbursed faster in the control mandals.
26 The dowry data also may suffer from underreporting since dowry payments have been illegal since 1961.
27 More rigorously, what we can identify by using the 2SLS estimation with an instrumental variable is so-called local average treatment effect (LATE), which is the average treatment effect among those who change their participation status by a change in the instrumental variable (Imbens and Angrist Citation1994).
28 The 2SLS estimation results show the average effect across all the households that were induced to participate in a SHG by the program placement in their residential mandals. Supplemental analysis reveals that the proportion of the wives who reported domestic violence was affected and the frequency of domestic violence among the wives who initially reported domestic violence was also affected by SHG participation.
29 William Chan (Citation2014) argues that dowry can also be interpreted as before death asset transfers from bride’s parents to the bride to help secure her welfare in the new household after marriage.
30 The inclusion of men in gender advocacy and gender training has long been recognized as necessary in programs that seek to promote gender equity and empowerment and has been found to have beneficial effects in SHGs (Koenig et al. Citation2003; Kim et al. Citation2007).
Additional information
Funding
Notes on contributors
Nozomi Sato
Nozomi Sato is Lecturer in the Department of Economics, Aichi Gakuin University. Her research interests lie in gender issues in South Asian countries, mainly Nepal and India. Her current research investigates women’s empowerment through the activities of women’s Self-Help Groups in India.
Yasuharu Shimamura
Yasuharu Shimamura is Associate Professor at the Graduate School of International Cooperation Studies, Kobe University. He is an economist who has carried out empirical studies in developing countries on various socioeconomic issues, particularly poverty, education, and health. He has experience working for international organizations such as the World Bank and the Japan International Cooperation Agency.
Susana Lastarria-Cornhiel
Susana Lastarria-Cornhiel is Emeritus Faculty at the Department of Planning and Landscape Architecture, University of Wisconsin–Madison. She is a sociologist who has worked on international development issues throughout her career with a focus on gender equity and rural property issues.