Although sanitation is essential for health, many households remain without a toilet due to financial constraints. This article describes the extent to which perceptions of costs and benefits of sanitation investments vary by gender and within the household. It demonstrates how these intra-household gender differences impact the success of a sanitation microcredit intervention in Maharashtra. Impacts of these differences on intervention success are largely concentrated in households where the woman is involved in the decision-making process.
Safe sanitation is recognised as being essential for primary health (“An adequate supply of safe water and basic sanitation” has been included in the Declaration of Alma-Ata). High rates of open defecation (OD) cause high mortality rates (Geruso and Spears 2018), and have long-lasing consequences on health and human capital accumulation (Miguel and Kremer 2004, Bleakley 2007, Augsburg and Rodriguez-Lesmes 2018, Spears, 2020). However, access to sanitation is not always straight forward. At the start of 2015, only 25% of households in our study area in rural Maharashtra owned a toilet – this is low even by rural Indian standards, where the toilet ownership rate is 56% (District Household Level Survey (DHLS) 2012-13).
Sanitation investment decisions
Toilet construction is a complex and costly process. A household that wishes to build a toilet must first identify and secure financial support or family savings. Then, household members would need to agree on the toilet model and location. Finally, someone would need to source the materials, hire skilled labour and oversee the construction. In rural India, where women’s movements outside the house are often restricted, these responsibilities often fall to male household members (Routray et al. 2017).
In our data, among households without a toilet, 83% report financial constraints as a key impediment to sanitation adoption. The average reported cost of toilet construction is Rs. 20,000, which is equivalent to around 50% of the average household annual income. However, financial tools and initiatives are often provided to women (as opposed to their husbands) as a way of promoting investment in goods and services believed to be preferred by women. For example, providing women with control over resources has been shown to improve female microenterprise investment (Riley 2020), female labour supply (Field et al. 2021) and nutrition (Armand et al. 2020, Duflo 2003).
Different household members may also have different perceptions on the costs and benefits of a toilet. One the one hand, women may perceive higher benefits from owning a toilet, as they find it more challenging to find a safe place to relieve themselves and are more subjected to sexual harassment. On the other hand, women who bear the brunt of the housework may also face the additional burden of fetching the water and cleaning the toilet. Consequently, it is likely that cost and benefit perceptions of both male and female household members matter in the sanitation investment process.
In a new study (Augsburg et al., 2021), we use novel data collected from both male and female adults in just over 1,100 households in rural Maharashtra and document the perceptions they hold regarding the costs and benefits of a standardised toilet, and highlight how these vary by gender and within the household. These households – who were all microfinance clients – were part of a cluster randomised trial studying the impacts of providing access to sanitation microloans. Thus, in a second step, we study how the intra-household gender differences in perceptions interact with female bargaining power to affect the success of the intervention1.
We collected information from one male and one female adult from each household. Firstly, individuals were asked to estimate the maximum and minimum amounts they would expect it to cost if they built the toilet in Figure 1.
Figure 1. Toilet model shown to respondents in perception elicitation exercise
From these estimates, we calculate an individual’s monetary cost perceptions as the mid-point of this minimum and maximum. In Figure 2 (left panel), we display the distribution of monetary cost perceptions separately for men and women. Overall, we see wide variation in expected monetary costs for this toilet. Although men and women perceive similar monetary costs, both tend to largely overestimate the cost of this toilet type. While the average reported cost of toilet construction in our survey is Rs. 20,000, the average expected monetary cost estimate is between Rs. 30,000 and 40,000.
Secondly, individuals were given a scenario household that had recently constructed the toilet shown. They were then asked to what extent they agreed with a series of statements relating to the non-monetary costs and benefits of toilet ownership2. In Figure 2 (centre panel), we show that women tend to perceive higher benefits to toilet ownership, on average, than men.
Finally, in Figure 2 (right panel), the distribution of non-monetary cost perceptions varies substantially by gender. Whilst the average male perceives relatively low non-monetary costs of toilet women fall into two groups – those who perceive higher costs and another group who perceive lower costs than the average male.
Figure 2. Distribution of individual perceptions of monetary cost (left), benefits (centre) and non-monetary cost (right) by gender
Collecting information from both male and female individuals enables us to compare differences in these perceptions within households (more specifically among married couples). As described in Figure 3, we are able to document substantial variation in intra-household gender differences for all three perceptions. Furthermore, for all perceptions, a large share of households have an intra-household difference that is equal to (or close to) zero which suggests that both male and female respondents have similar perceptions. In Figure 3 (right panel), which shows the distribution of intra-household benefit perceptions, we see that women have a higher benefit perception than men in a significant proportion of households.
Figure 3. Distribution of intra-household perceptions of monetary cost (left), benefits (centre) and non-monetary cost (right)
We study these perceptions in the context of a sanitation loan intervention carried out between February 2015 and September 2017 in rural Maharashtra. The intervention made new sanitation loans available to existing female clients of a large microfinance institution (MFI) for the construction of a new toilet or the repair and upgrade of an existing toilet. However, no information or advice on toilet construction was provided by the MFI. The loans were provided collateral-free to women in joint-liability groups3. Receipt of the loan required agreement from the spouse and repayments were made weekly over a two-year period. The loans carried an average annual interest rate of 20% (which was lower than other commonly available loans such as business loans). The new loans were made available to microfinance clients in 40 out of 81 randomly selected gram panchayats (GPs). In the remaining 41 GPs all other services continued as normal.
We use the randomised setting provided by the sanitation loan intervention to investigate how individual perception and decision-making power affects uptake of the loan and sanitation investments. To guide the empirical analysis, we develop a theoretical model of household decision-making in which a couple bargains over borrowing, investment and consumption choices. The model predicts that improved access to credit has a larger effect on loan uptake and investments when the household member with less decision-making power has the higher net perceived benefit from the investment.
We first estimate how the effect of the intervention on loan uptake and sanitation ownership differs by intra-household differences in perceptions. We find that these differences do matter, but at different stages of the process. As shown in Figure 4 (left panel), households where the woman perceives a higher or similar benefit compared to her husband are four times more likely to take a sanitation loan compared to households where the woman’s benefit perception is lower. By contrast, other perceptions do not have differential impacts on sanitation loan uptake. In Figure 4 (right panel) we see treatment effects on toilet ownership are significantly higher in households where the woman has a lower cost perception than the man.
Figure 4. Differential treatment effects by intra-household perceptions on loan uptake (left) and toilet ownership (right)
We further break down the effects by the woman’s decision-making power. Our measure of decision-making power is based on questions asked to the female respondent about who makes the final decision on purchasing large household items, household expenditures and spending money received by women4. While the distribution of this measure does not, on its own, generate differential impacts, the differential impacts we saw in Figure 4 are observed only in households where the woman has a mid-level decision-making power. These are women who are often involved in making household decisions, but alongside at least one other household member.
We show these results in Figure 5 (left panel) which breaks down the treatment effect on loan uptake by intra-household benefit perceptions (higher or lower) and the wife’s level of bargaining power, and in Figure 5 (right panel) which shows the breakdown of the treatment effect on toilet ownership by the intra-household monetary cost perception difference and the wife’s level of bargaining power. When the wife has similar or higher benefit perception compared to her husband, and is in the mid-bargaining power (BP) sub-group, household loan take-up is 32 percentage points higher than for households where the woman has mid-BP and a lower perception of the benefits than her husband. Furthermore, toilet ownership in households where the wife has a lower perception of monetary costs and mid-BP increases by 30 percentage points which is significantly higher than for those where the wife has similar or higher monetary cost perception and mid-BP.
Figure 5. Differential treatment effect by wife’s bargaining power and intra-household benefit perception (left) and intra-household monetary cost perception (right)
Conclusion and policy implications
Our findings conclude that intra-household differences in perceptions only influence borrowing and investment where the woman has some involvement in decision-making. Targeting sanitation microcredit to women can be effective in increasing sanitation investments. However, we highlight the importance of combining these interventions with information, not just on the benefits of sanitation ownership but also the monetary costs to address the significant misperceptions in monetary costs, especially as our evidence suggests this would increase the conversion of loans to toilets.
- Augsburg et al. (2022) study the average impacts of the intervention and report that 20% of clients took a sanitation loan as a result of the intervention, leading to a 9 percentage point increase in toilet ownership.
- Benefit statements related to positive health benefits, family status, women’s safety and time savings. Non-monetary cost statements related to negative health outcomes, time spent fetching water and loss of exercise and social time.
- Under joint-liability, each group member is also responsible for the repayment of loans taken by her group members. If one member were to default, all other members would be responsible for repaying the defaulting member’s loan.
- The respondent can include multiple decision makers, including in-laws.
- Armand, Alex, Orazio Attanasio, Pedro Carneiro and Valerie Lechene (2020), “The Effect of Gender-Targeted Conditional Cash Transfers on Household Expenditures: Evidence from a Randomized Experiment”, The Economic Journal, 130(631): 1875-1897.
- Augsburg, B, B Malde, H Olorenshaw and Z Wahhaj (2021), ‘To invest or not to invest in sanitation: The role of intra-household gender differences in perceptions and bargaining power’, IFS Working Paper W21/45.
- Augsburg, Britta and Paul Andres Rodríguez-Lesmes (2018), “Sanitation and child health in India”, World Development, 107: 22-39.
- Augsburg, B, S Smets, B Malde, S Giunti and B Caeyers (2022), ‘Sanitation and credit constraints – The role of labelled microcredit in India’s Swacch Bharat Mission subsidy scheme’, IFS Working Paper W22/09.
- Bleakley, Hoyt (2007), “Disease and Development: Evidence from Hookworm Eradication in the American South”, The Quarterly Journal of Economics, 122(1): 73-117.
- Duflo, Esther (2003), “Grandmothers and Granddaughters: Old-Age Pensions and Intrahousehold Allocation in South Africa”, The World Bank Economic Review, 17(1): 1-25.
- Field, Erica, Rohini Pande, Natalia Rigol, Simone Schaner and Charity Troyer Moore (2021), “On Her Own Account: How Strengthening Women’s Financial Control Impacts Labor Supply and Gender Norms”, American Economic Review, 111(7): 2342-2375.
- Geruso, Michael and Dean Spears (2018), “Neighborhood Sanitation and Infant Mortality”, American Economic Journal: Applied Economics, 10(2): 125-162.
- Miguel, Edward and Michael Kremer (2004), “Worms: Identifying Impacts on Education and Health in the Presence of Treatment Externalities”, Econometrica, 72(1): 159-217.
- Riley, E (2022), ‘Resisting social pressure in the household using mobile money: Experimental evidence on microenterprise investment in Uganda’, CSAE Working Paper WPS/2022-04.
- Routray, Parimita, Belen Torondel, Thomas Clasen and Wolf-Peter Schmidt (2017), “Women’s role in sanitation decision making in rural coastal Odisha, India”, PLoS ONE, 12(5): 1-17.