Getting India's women into the workforce: Time for a smart approach
- 10 March, 2017
In this article, Rohini Pande, the Mohammed Kamal Professor of Public Policy at Harvard Kennedy School, contends that raising India's stubbornly low rate of female labour force participation will require behavioural interventions that address social norms.
Tweet using #womenandwork
Between 1990 and 2015, India's real GDP (gross domestic product) per capita grew from US$375 to US$1572, but its female labour force participation rate (LFPR) fell from 37% to 28%. This gives us a puzzle to solve: why isn't India following the same trajectory as most other countries at a similar level of growth, where female LFP rises with GDP?
The answer isn't a lack of interest on the part of women. The 2011 National Sample Survey (NSS) shows that over a third of women engaged primarily in housework say they would like a job. That number rises to close to half among the most educated women in rural India.
The answer isn't lack of apparent political will, either. The Government of India has poured huge resources into girls' schooling, but that hasn't translated into higher female LFP as it has in other countries (Heath and Jayachandran 2016). Initiatives such as 'Skill India' and 'Make in India' include quotas to ensure a certain proportion of trainees are women. Yet, programmes struggle to recruit women, place them in jobs, and keep them in jobs once they are placed.
Instead, a growing body of research suggests that an important part of the answer appears to be social norms about appropriate behaviour for women, and the enforcement of these norms by women's parents, husbands, and parents-in-law. Take mobility, for instance. The 2011 Indian Human Development Survey (IHDS) shows that a very sizeable fraction of Indian women say that they require permission from a family member even to go to the local market or health centre. In the end, it's pretty difficult to look for a job if you can't leave the house alone.
The ubiquitous role of the family in enforcing these norms also shows up in the reasons why women are often unable to hold down a job. A survey of men and women who entered skills training programmes shows that family pressures and responsibilities was by far the most common reason women didn't accept jobs or quit them, while low pay was the main reason for men.
Raising female LFP matters, not least because many Indian women want to work. These preferences likely reflect the greater autonomy enjoyed by working women. Research, for instance, finds that girls who have the option to work – or even who have an older sister who works – marry later (Sivasankaran 2014). And later marriage, in turn, delays child-bearing, which is associated with health benefits (Field and Ambrus 2008). And, finally, it is very likely that an economy grows faster when all the most able citizens, irrespective of gender, can enter the labour force (Hsieh et al. 2013).
When the stumbling block is social norms rather than, say, lack of resources or human capital, then it changes how we can make progress and calls for a smarter policy response. Below, I lay out some elements of what, I think, this would entail.
Get the measurements right
The impact of policies and private-sector initiatives on women's empowerment is a tricky thing to measure, so we tend to use proxies such as the answer to particular survey questions. However, focussing on the wrong proxies can cause us to misidentify programme impact. In a repressive society, a woman must reject powerful social norms to claim empowerment. It is critical that we recognise that the causal chain linking a programme to women's ability to successfully challenge social norms has many links. Simply focussing on the last one – whether a norm was successfully challenged – may cause us to underestimate the value of several public policies.
I'll give an example. The January 2015 issue of American Economic Journal: Applied Economics represented a landmark in the study of microfinance. Six papers based on large-scale randomised trials in six different countries across four continents showed that the poverty-reduction effects of microloans were far more modest than the industry had claimed. Four of the six studies also estimated the effect on women's empowerment, and three of these found "no evidence of transformative effects on social indicators."
Here, the proxy for empowerment was women's self-described decision-making power in the household. Arguably, more decision-making power within the household reflects greater bargaining power and an ability to successfully challenge the norm that the husband's preferences are paramount. The last may, however, take a while to occur. And getting to the point where a woman, who still lives in a socially conservative environment, will tell a surveyor the ways she has so challenged conventions, especially if her husband or other members of his family are in the room, will likely take even longer. It may be easier for a woman to describe the work she does and the loans she takes out, rather than the struggle she faced to make that possible.
Indeed, if the authors had dwelled on their findings concerning women's work habits they may have concluded differently. In India – arguably the most conservative of the three countries where female-only lending was analysed – total household labour supply went up, and did so through self-employment activities, which may well reflect that women worked more. Yet, women's reported decision-making power did not improve.
In India, self-employment is the avenue most women use as they move out of agriculture into other work, so shifting Indian women's time use toward their microbusiness should be seen as a success. We shouldn't wait for a woman's self-reported decision-making power to change. That might come much later, and her economic empowerment might improve in the meantime, along with many benefits that correlate with financial independence. Women state that they would like to work; women's work correlates with a number of empowerment measures, and so if microfinance encourages work or empowers women financially enough to make decisions about work – that may be valuable, in itself, even without making that final step of whether decision-making behaviours change.
Get the policies right
So, the goal should be to provide economic opportunities for women who want work, and allow the benefits to accrue. But how do we do this when norms militate against women's work? Let me describe two insights from my recent work with colleagues that give hints on how to change women's behaviour and draw women into the workforce.
Foster women's social networks
In a recent study (Pande et al. 2015), my co-authors and I worked with Self-Employed Women's Association (SEWA) Bank, the largest women's bank in India, to offer women customers a two-day course on business skills and how to set goals. In the treatment group (those who were offered the course), half of the women received the offer for them alone, and the other half were invited to bring a friend. We found that, while both sets of women were more likely later to take out a loan than women who didn't take the course, those who had a friend along were even more likely. Furthermore, those who came with a friend were more likely to use their loans specifically for business purposes, while the women who were invited alone mostly used them for home repair - largely unrelated to their businesses.
Most strikingly, women who were invited to the training with a friend had significantly higher household income and consumption levels four months later, and were less likely to report their occupation as housewife.
The larger lesson is that when women have little bargaining power in the household, they need to revert to their networks with other women to effect change: we've seen this happen in the temperance movement, where women cooperate to influence a household decision, that is, whether their husbands drink. Women can tap into that power to influence other household decisions – what kind of investments to make, and whether or not to work.
Give women control over the money they earn
India's rural workfare programme MNREGA (Mahatma Gandhi National Rural Employment Guarantee Act) has achieved some success in targeting women, in part because it provides them employment close to home. But, until recently, the default was to pay wages into the bank account of the head-of-household, usually male. My colleagues and I ran a large-scale experiment in Madhya Pradesh to test the effect of several combinations of interventions: providing them with their own bank account, having their MNREGA payments directly transferred there, and giving them basic financial literacy training so they know how to use the account. The study is still underway, but we already have some remarkable results.
We found a range of financial benefits from the different interventions, but women who received all three showed the biggest changes: they were more likely to have worked more, both under and outside MNREGA; they reported 25% higher earnings and 60% higher bank balances, and were more likely to make household purchases with their own money.
India may have a lot to learn from its poorer neighbours when it comes to female LFP. While across South Asia, men participate in the labour force at the same rate of around 80%, women's participation varies widely. Pakistan's female LFPR is lower than India's at 25%, but it is rising, while in Bangladesh, which has similar social norms as in India, female LFPR has hovered around 60% for decades.
What is Pakistan doing right? How did Bangladesh overcome restrictions on women's mobility and attitudes regarding their 'purity', to allow them to participate in the garment industry in large numbers? On 23-24 March 2017, the research group I co-direct, Evidence for Policy Design (EPoD), in collaboration with International Growth Centre (IGC) and Institute for Financial Management and Research (IFMR LEAD), will bring together international scholars and policy actors from across South Asia in Kathmandu, Nepal, to share research on how government initiatives to increase financial access and economic opportunities are affecting the region's women.
- Field, Erica and Attila Ambrus (2008), 'Early marriage, age of menarche, and female schooling attainment in Bangladesh', Journal of Political Economy, 116(5): 881-930. Available here.
- Field, E, S Jayachandran, R Pande and N Rigol (2015), 'Friendship at Work: Can Peer Effects Catalyze Female Entrepreneurship?', Harvard Kennedy School, Working Paper Series 15-019.
- Heath, R and S Jayachandran (2016), 'The Causes and Consequences of Increased Female Education and Labor Force Participation in Developing Countries', National Bureau of Economic Research (NBER), Working Paper No. 22766.
- Hsieh, C, E Hurst, CI Jones and PJ Klenow (2013), 'The Allocation of Talent and US Economic Growth', National Bureau of Economic Research (NBER), Working Paper No. 18693.
- Sivasankaran, A (2014), 'Work and Women's Marriage, Fertility, and Empowerment: Evidence from Textile Mill Employment in India', Job Market Paper.