This is the 11th in this year’s series of posts by PhD students on the job market.
Why are so many women willing to work but unemployed? This question has inspired many development programs, most of which target women directly. Yet women must negotiate work decisions within households where men are often the key decision-makers - a critical piece of the low female employment puzzle.
In my job market paper, I study the preferences and expectations of spouses that factor into household decisions about female labor force participation in India. Novel survey data and a field experiment reveal that men expect employed wives to gain decision-making power, and this perceived reallocation of resources imposes costs on them. Preserving their role as decision-makers creates an incentive to suppress women’s employment (Basu, 2006). There is an upside: if men’s beliefs could be changed to match their wives’, female employment would increase substantially.
The Setting: Couples in Uttar Pradesh
The study took place in Lucknow, Uttar Pradesh, where I surveyed both spouses from 1,100 young married couples. The desire to work among women is ubiquitous: 92% of surveyed women want to work. Yet, only 26% are currently employed. Most wives report limited agency: in 75% of couples, men have the final say on whether the wife accepts a job offer.
I use a combination of survey experiments and hypothetical scenarios to measure household outcomes and individual expectations over how the allocation of resources of spouses’ time would change if the wife were to join the labor force. Matched data from both spouses provides a rare window into the mental models driving household decisions on female market participation.
Husbands and wives’ beliefs align along two dimensions. First, female employment does not lead to a reallocation of chores and care duties which are, and are expected to remain, the wife’s primary responsibility – the so-called second shift. However, both spouse’s agree that a woman who works for pay has more say over household’s spending decisions. The key difference is in the magnitude of these expected changes across genders.
Divergent expectations
How does this expected shift in decision-making power over spending differ across spouses? I measure this by sequentially showing respondents five scenarios, one in which the wife dedicates all of her time to home production, and four where she is employed, varying her hours and wage but keeping the husband’s labor market outcomes constant. Wives and husbands are then asked how their household would allocate their disposable income in each scenario, split between joint expenses – food, expenditure on children, etc. – and the private consumption of each spouse.
The results are shown in Figure 1. The horizontal axis represents the wife’s hypothetical income relative to her husband, while the vertical axis represents her private consumption relative to his. Regardless of current employment status, women expect earning more would modestly increase their share of private consumption relative to their husband. However, even if they were to earn 50% more than their spouse, they still expect to consume less than he does.
A stark difference emerges from men’s answers, especially those from single-income couples (where only the husband earns income). In the scenario that is most common among working women, that in which the wife earns half as much as her husband, breadwinning men expect their spouse to achieve an equal share of private consumption, and to get more than him if she were to earn the same or more than he does. Moreover, these scenarios are not unlikely: among the four jobs presented to them, men believe there is a 46% chance that their wife could find one where she earns as much or more than he does.
Also striking is the difference between the expectations of this group, and those held by men whose wife is currently working for pay. The latter expect to remain the primary consumer as long as they are the primary earner - as almost all of them are – and this role will be challenged only if the wife were to earn substantially more than they do (cross the vertical line of equal shares).
Small changes, large costs
These beliefs matter because they affect decisions. The paper includes a theoretical framework that makes this link precise, but the intuition is simple: even if the wife’s labor income would improve family welfare, the husband may prefer the status quo to the expectation of losing control over financial decisions or a lower relative position in terms of private consumption.
I measure spouses’ preferences over how the household is organized through a discrete choice experiment. Respondents are shown pairs of vignettes depicting different household profiles. These vary total income available to the household, but also how resources, and the burden of work and home production are allocated between spouse.
Men show a consistently large distaste towards women being the primary spender, even if the household as a whole is richer. The survey experiment allows me to quantify this distaste: the median husband, earning INR 450 per day, would need INR 250 (+56%) of extra household income to be compensated for his wife getting a higher share of private consumption than he does, whether she works or not. This cost is distinct from the social norms that surround women’s work, and compounds them, adding another barrier to female participation in labor markets.
Testing the mechanism: A Job offer experiment
To test whether these preferences and expectations truly constrain women’s work, I offer women in the sample a data entry job, randomizing whether the offer is communicated directly to the wife or to her husband. This variation mimics reality: homebound women often hear about jobs through male relatives. Importantly, I stratify by whether the husband expects a job such as this one, if permanent, would give his wife a greater than 50% share of private consumption (henceforth pessimistic expectations).
When wives were informed directly, their likelihood of accepting the job depended heavily on their husbands’ prior beliefs. Women who had to negotiate participation with husbands holding pessimistic expectations (that his share would be less than half of private consumption) were 40% less likely to take up the job (see Figure 2).
When the husband was told first, instead, job take-up is much lower. Interestingly, take-up does not vary by his expectations in this case, and this finding teaches us something about the mechanism that keeps women out of the labor force. In couples where the man holds pessimistic beliefs, take-up is low regardless of who receives the offer – men’s opposition in this case is likely too strong.
The lower average take-up when the offer is communicated to men, and the lack of differences in take-up by their expectations, is entirely driven by those who hold moderate expectations and whose wife is not currently working. Their wife would have become employed had she received the offer directly, but these men appear to gatekeep the status quo.
Aligning expectations to unlock women’s work
This study quantifies how men’s expectations and preferences shape women’s labor force participation decisions, and shows that these expectations are systematically pessimistic. They anticipate large personal costs from their wives’ employment, fearing a loss of decision-making power and control over household resources. These perceived costs reinforce existing social norms that already discourage women’s work. Yet the findings also point to an actionable opportunity: if men’s expectations were aligned with those of their wives—who foresee only modest changes—female labor force participation would rise by 5 percentage points, a 19% increase over baseline levels. While long-run change will require broader shifts in gender norms and labor market policies, short-run interventions that address men’s beliefs and household decision-making could offer a feasible and effective path toward unlocking women’s economic potential.
Andrea Smurra is a post-doc at the London School of Economics and Political Science.
Join the Conversation