Do Traditional Institutions Constrain Female

Entrepreneurship? A Field Experiment

on Business Training in India
By Erica Field and Seema Jayachandran and Rohini Pande

What constrains the entrepreneurial choices and caste institutions that constrain womens
of poor women? Do traditional institutions pose mobility and behavior influence their business
unique barriers to business growth and prof- activity.
itability for female-run enterprises? Our analysis makes use of a field experiment
The explosion of microfinance programs, in which a randomly selected sample of poor self-
which typically target this group, has drawn at- employed women were trained in basic financial
tention to these questions. Indeed, one view is literacy and business skills and encouraged to
that inadequate access to credit prevents women identify medium-term financial goals. The sam-
from undertaking high-return business activi- ple is relatively homogenous in terms of socioe-
ties. However, the empirical evidence is mixed. conomic status (e.g., education). However, these
For example, Suresh De Mel, David McKenzie women differ in religion and caste and, there-
and Christopher Woodruff (2008) find low re- fore, face very different traditional restrictions
turns to capital in female-run enterprises. on mobility and social interactions. Muslim
Thus, another view is that the main con- women face the most restrictions. Among Hindu
straint is limited demand for rather than supply women, upper castes (hereafter, UC) face signif-
of credit, with poor women lacking high-return icantly more restrictions than scheduled castes
ways to expand their businesses. Here, a first (hereafter, SC), the lowest group in the caste
possibility is lack of knowledge: Women might hierarchy.
be relatively uninformed about investment op-
In general, the returns to entrepreneurship
portunities and untrained in basic cost-benefit
should be highest for the women least fettered by
analysis that could improve their profitability.
conservative social norms. However, this need
A second possibility is that norms govern-
not be the case for an intervention that primarily
ing womens roles in society limit womens per-
influences womens knowledge of business prac-
ceptions about what is achievable in the work-
tices and their aspirations. If traditional norms
place.1 Even the knowledge hypothesis given
about gender roles can be challenged, or if they
above might have as its root cause social norms
mainly work to limit womens exposure to and
about what girls and women are taught.
knowledge of business opportunities, then the
This paper explores how traditional religious
returns from training may be higher for women
Field: Harvard University, [email protected].
from more restrictive social groups.
edu. Jayachandran: Stanford University, jay- Our results provide some support for this the-
[email protected]. Pande: Harvard University, sis: Among Hindu women, training increased
rohini [email protected]. We are grateful to Man- borrowing and business income for those facing
asee Desai, Katherine Durlacher, Sarah Moshary, more restrictions, i.e UC women. However, Mus-
Natalia Rigol, Mallika Thomas, and Divya Varma lim women failed to benefit from the training
for excellent research assistance, the staff of SEWA program. We interpret these patterns as sug-
for collaboration and the Center for Microfinance gestive of a non-monotonic relationship between
and ICICI for logistic and financial support.
1 Another possibility is that women are unable to social restrictions and the ability to benefit from
pursue high-return activities because, low bargaining business training. Consistent with this interpre-
power vis-a-vis their husbands, limits their control tation, we also observe a non-monotonic effect
over how their money is spent. of restrictedness within the broad social groups.
1
2 PAPERS AND PROCEEDINGS MAY 2010

I. Gender norms in India count, and roughly a quarter of clients have ever
taken out a loan from SEWA Bank.
The Hindu caste system developed as an ex- For several years, SEWA Bank has run a five-
tremely hierarchical social system. Its defining day training program on financial literacy. The
principles include the ideas of purity and pollu- curriculum, developed by Freedom from Hunger
tion. UCs maintain purity by avoiding sexual re- and used widely around the world, teaches ba-
lations, marriage and, in extreme cases, contact sic accounting skills and about interest rates
with lower castes (especially SCs). Premised on and life-cycle planning, among other topics. It
men being a source of pollution for women, re- emphasizes financial prudence and encourages
strictions are placed on women to limit their women to avoid excess debt, save more and re-
contact with men other than husbands. The duce frivolous spending (for example, on tea).
requirements include that a married woman re- More recently, SEWA Bank started a second
main veiled, not interact with older men, have five-day course that teaches business skills such
restricted mobility outside of her husbands as marketing, cost reduction, investment, and
house, and not remarry if widowed. One effect customer service.
of these norms is that they significantly restrict In collaboration with SEWA Bank, we de-
female labor force participation. signed a streamlined two-day training module
Within the caste system, concerns of pu- that combined elements of the existing financial
rity and pollution are less relevant for SCs, literacy and business skills curricula, and added
who rank low in the hierarchy. In addi- new material focused on aspirations. The as-
tion, greater poverty implies increased reliance pirations component included a short film show-
on female wage-earning (M. Acharya and L. casing successful SEWA members who have used
Bennett 1983, Karin Kapadia 1997). As a result, good financial practices to bring themselves out
SC women face fewer social restrictions and, of poverty. As homework after the first day
by virtue of being independent earners, enjoy of training, the women had to fill out a work-
greater financially independence from their hus- sheet identifying a financial goal they wanted
bands and increased control over household fi- to achieve over the next six months. The sec-
nancial decisions relative to UC women (Joan P. ond day of training, the participants finished the
Mencher 1988). Notably, these restrictions on worksheet, breaking down their goal into smaller
female autonomy among UCs are not limited short-run steps.
to the wealthy (Mukesh Eswaran, Bharat Ra- For the experiment, 636 women were ran-
maswami and Wilima Wadhwa 2009). domly drawn from the SEWA Bank customer
Relative to Hindus, Muslims in India place database who met three criteria: actively sav-
more restrictions on womens contact with those ing or borrowing from SEWA Bank in the past
outside the sphere of kinship. Because Muslim two years, owns a business or is self-employed
women are entitled to a share in the family real and is age 18 to 50. Two thirds of the women
estate, controlling their relationships with males in this sample were then randomly assigned to
outside the family can be crucial to the mainte- the treatment group, in which case they were re-
nance of family property and prestige. cruited for business training; the control group
was not invited to attend training.2
II. Intervention and study design Women in the treatment group were recruited
to attend a particular training session at their
We conducted a business training interven- nearest SEWA branch. Home visits were used
tion in conjunction with SEWA Bank, which to invite eight women per training session. For
is based in the city of Ahmedabad in western data collection and analysis purposes, women in
India. Its 170,000 member-clients are primar- the control group were also assigned but not re-
ily poor women who work in the informal sec- cruited to a particular training session at their
tor (for example as incense-stick makers, tailors,
and vegetable vendors). SEWA Bank offers 2 The randomization was stratified by whether
these women a wide array of financial products. the individual was part of the first or second phase
All clients are required to have a savings ac- of sampling and by SEWA branch.
 VOL. 100 NO. 2 TRADITIONAL INSTITUTIONS AND FEMALE ENTREPRENEURSHIP 3

nearest SEWA location, allowing us to cluster was high, with over 70% of those invited choos-
standard errors by training session. ing to attend. SCs, who face fewer restrictions
Our analysis sample comprises the 597 women that UC and Muslims, were nearly one third
who were successfully surveyed at follow-up and more likely to attend the training than Mus-
could be categorized into sub-castes (based on lims and UCs. The results on take-up, which
their surname).3 are also the first stage of our treatment-on-the-
We categorized women into three broad so- treated (TOT) results, are reported in the online
cial groups: Muslim, Hindu SCs, and Hindu appendix.
UCs (non-scheduled castes and other backward Baseline characteristics are balanced across
castes). In addition, we scored (from 1 to 4) how the control and treatment groups (results re-
restrictive each Hindu and Muslim sub-caste was ported in the online appendix). Table 1
in regard to five norms governing womens be- compares baseline characteristics across social
havior: ability to socialize alone, requirements groups. Women in our sample are strikingly ho-
to cover the face or wear a veil, ability to speak mogenous across social groups: Average educa-
directly to elders, ability to leave the house or tion and family size are almost identical for Mus-
neighborhood alone, and ability to remarry. We lims, UCs and SCs, and SC women have higher
created an index equal to the number of norms household income and are slightly more likely to
for which the sub-caste was highly restrictive own a family business, though the differences are
(score of 4); thus, the index runs from 0 to 5. statistically insignificant. However, the groups
0.95 Figure 1 shows the proportion of women sub- show large differences in social restrictions. UCs
0.31 ject to a highly restrictive norm about veiling. are more restricted than SCs, and the difference
0.18 SC women face the least restrictions, followed by for Muslim women is particularly strikingthey
UCs and Muslim women. This pattern is true are more than twice as likely to face severe re-
for all norms, apart from interaction with elders strictions as UCs and over three times as likely
(where Muslims face less restrictions). as SCs.

1 0.95 Table 1Baseline characteristics

0.8 UC SC Muslim
Education 6.33 6.62 6.36
0.6 Income 4852.6 5694.6 5189.5
Household size 5.25 5.25 5.42
0.4 0.31
Owns business 0.27 0.31 0.26
0.18
0.2 Restricted 0.75 0.48 1.70***
N 346 70 181
0
Notes: Asterisks denote significant differences in
Muslims Upper Scheduled means between the group and UCs. In all tables *
caste caste denotes significant at 5% level ** significant at 1%
Hindus Hindus level *** significant at 0.1% level.

Figure 1. Proportion with veiling as the social

norm

III. Effects of business training

The intervention trained 289 women in
57 two-day training sessions conducted from Our estimation strategy exploits the random
September 2006 to April 2007. Program take-up assignment to treatment, i.e. to being invited
to a training session. We examine the impact
3 The survey attrition rate (32 of 604) is similar of attending the training program on economic
across treatment and control groups. We were un- outcomes, instrumenting for attendance with
able to assign caste to seven cases. whether the participant was in the treatment
4 PAPERS AND PROCEEDINGS MAY 2010

group. This IV specification provides TOT esti- sured in rupees, across treatment and control for
mates. We separate out the differential effect of any of the groups, though the point estimates
training by social group by interacting the train- again go in opposite directions for UCs com-
ing dummy with indicators for being a SC and pared to SCs and Muslims, with UCs appearing
a Muslim. Our outcome variables are from a to substitute away from saving. This pattern is
survey conducted on a rolling basis four months consistent with their increased borrowing.
after training. To more directly measure the effect of train-
The results in Table 2 reveal that the train- ing on business activity, in Table 3 we exam-
ing led to a significant increase (13 percentage ine the womens personal business income, both
points) in the likelihood of taking out a loan one the total amount over the past week measured
to four months after the training (Borrowed) in rupees (Income) and an indicator of whether
among UCs, who are the omitted category. UCs she reports any personal business income (Any
who attended the training took out loans at Income). Although we sampled on being em-
nearly twice the rate as UCs in the control ployed, many of the women do not report regu-
group. Meanwhile, we cannot reject that there lar earnings; in the control group, only 80% of
was no effect of the training on borrowing among women report any earning in the past week.
either SC or Muslim women. We observe a positive and significant effect
When we estimate the training effect on of the training on business income among UCs
the likelihood that a woman reported problems only, suggesting that the new loans were put to-
managing her debt, we find no overall effect and ward business investments. The point estimates
no differential effects across social groups (re- on business income suggest an increase of around
sults not reported). Administrative bank data 30%, but the results are too imprecise to draw
on loan default confirm this result. Thus, the conclusions. The estimated effect of training on
training does not seem to have induced UC earning any business income is highly significant
women to borrow beyond their means. among UCs, indicating a 25% increase in the
likelihood that a woman engages in labor market
Table 2Treatment effects on borrowing and activity. Together with the loan results, this pat-
saving. tern suggests that the training encouraged UCs
to start or expand their microenterprises. Fur-
Borrowed Savings ther supporting these results, we find that train-
Trained 0.13* -315.32 ing led UCs but not other women to talk more
(0.07) (492.83) frequently with family members about business
TrainedSC -0.16 444.71 plans (results not reported).
(0.14) (540.02)
TrainedMuslim -0.14 317.51
(0.11) (630.07) Table 3Treatment effects on income.
SC 0.04 -298.12
(0.10) (301.87) Income Any Income
Muslim 0.04 -46.50 Trained 273.3 0.19**
(0.06) (333.99) (228.0) (0.09)
Mean of dep. var. 0.17 277.59 TrainedSC -525.4 -0.37***
(350.2) (0.14)
Notes: Standard errors clustered by training TrainedMuslim -545.2* -0.22*
session. Regressions include SEWA branch, (313.9) (0.12)
treatment month, and sampling phase fixed effects. SC 290.7 0.25***
Mean of dependent variable is for UCs in the (183.9) (0.08)
control group. N=597. See online appendix for Muslim 109.2 0.09
further details. These notes also apply to Table 3. (170.8) (0.07)
Mean of dep. var. 858.4 0.79

There are no significant differences in savings

during the past month (Savings), which is mea-
VOL. 100 NO. 2 TRADITIONAL INSTITUTIONS AND FEMALE ENTREPRENEURSHIP 5

IV. Discussion selves of entrepreneurial opportunities. We find

evidence consistent with restrictedness having
Given the similarity in household wealth and a non-monotonic effect when we examine het-
types of businesses across social groups, the dif- erogeneous treatment effects by restrictedness
ference across the groups in their response to within social groups (see the online appendix).
business training is stark. It is made even more Our business counseling program significantly
striking by the fact that data collected during reduced the business income gap between social
the training reveal identical patterns of business groups. Thus, another way of reading our results
and financial goals across social groups. is thatin the absence of interventions that
counteract traditional normsmodernization is
One possibility is that differences in treat-
likely to yield greater benefits for women lower
ment effects are driven by lower program take-up
in the caste hierarchy, a point also made by
among UCs. The training may have attracted
Kaivan Munshi and Mark R Rosenzweig (2006).
a highly selected sample of UCs who were es-
This view, however, assumes that social norms
pecially responsive to training. While there is
for lower castes continue to be less restrictive.
higher baseline business ownership among UC
If, instead, modernization heightens sanskritiza-
trainees relative to other trainees, the treat-
tionthe desire of lower castes to emulate upper
ment differences persist even when we control
castes and move up the caste hierarchywith
for this (and/or other) observable characteris-
SCs increasingly adopting the gender norms of
tics. Thus, heterogeneity in take-up does not
UCs, then economic growth may fail to emanci-
seem to explain the heterogeneous treatment ef-
pate women to the same extent.
fects (though we cannot rule out unobservable
differences across groups).
REFERENCES
If the explanation for UC women being es-
pecially responsive to training is that social re- Acharya, M., and L. Bennett. 1983.
strictions created knowledge deficits or the train- Women and the Subsistence Sector: Economic
ing allowed them to challenge social norms that Participation and Household Decision Making
were distorting their business practices, then an in Nepal. World Bank Working Paper No. 526.
important question is why Muslim women, who De Mel, Suresh, David McKenzie, and
face the highest degree of restrictions, did not re- Christopher Woodruff. 2008. Returns to
spond more to the training than SC women did. Capital in Microenterprises: Evidence from a
One possibility is that, although restrictions are Field Experiment. Quarterly Journal of Eco-
greater for the average Muslim woman compared nomics, 123(4): 13291372.
to the average SC woman in Ahmedabad, there
Eswaran, Mukesh, Bharat Ramaswami,
is little difference in restrictions across Muslim
and Wilima Wadhwa. 2009. Status, Caste,
and SC members of SEWA Bank. Unfortu-
and the Time Allocation of Women in Rural In-
nately, without individual-level data on restrict-
dia. mimeo, University of British Columbia.
edness, we cannot test this story. Another pos-
sibility is that Muslims in Ahmedabad, which Kapadia, Karin. 1997. Mediating the Mean-
has a history of religious tension, face consid- ing of Market Opportunities: Gender, Caste
erable discrimination in the marketplace, which and Class in Rural South India. Economic and
business training could not undo. Political Weekly, 35(52): 33293335.
Bearing in mind these caveats, a prima facie Mencher, Joan P. 1988. Womens Work and
explanation for our results is non-monotonicity Poverty: Womens Contribution to Household
in the effect of social restrictions: The train- Maintenance in South India. In A Home Di-
ing helped women whose businesses had been vided. , ed. Daisy Dwyer and Judith Bruce.
held down by social restrictions, but women Stanford, CA:Stanford University Press.
subject to extreme restrictions had too little Munshi, Kaivan, and Mark R Rosen-
agency to easily change their aspirations or ac- zweig. 2006. Traditional Institutions Meet
tivities. Even with more knowledge or higher the Modern World: Caste, Gender and School-
aspirations, the most restricted women might ing Choice in a Globalizing Economy. Ameri-
face too many social strictures to avail them- can Economic Review, 96(4): 12251252.
6 PAPERS AND PROCEEDINGS MAY 2010

Online Appendix

Table A.1First stage.

Trained Trained
Treated 0.696*** 0.654***
(0.0266) (0.0368)
TreatedSC 0.180***
(0.0603)
TreatedMuslim 0.0711
(0.0608)
SC -0.0199
(0.0184)
Muslim 0.0466**
(0.0229)
N 597 597