The approval we imagine

FOR a quarter of a century, nearly everything that is supposed to bring women into the workforce has been moving in the right direction in Pakistan. Female university enrolment has grown roughly six-fold since 2000. Fertility has fallen, and families continue to move from villages to cities, where jobs are closer and more varied. In the standard economic playbook, these are the three important engines of women’s employment: education raises the wages a woman can earn, fewer children free her time, and urban labour markets put work within reach. And yet female labour force participation has barely moved. It has sat in the 22-25 per cent range for years, among the lowest in the world, while roughly four out of five working-age men are in the labour force. The engines are running but the vehicle is stationary.

This is what makes Pakistan an exception, and a puzzle. In most countries that experienced comparable gains in women’s education and comparable declines in fertility, women’s employment rose. Bangladesh’s garment boom transformed the lives of millions of women whose profiles were not so different from those of women in Pakistan. Pakistan has similar garment factories, often recruiting from the same kind of neighbourhoods, and the surge never came. Models based on preferences and material constraints would predict higher participation than we observe, which means something else is holding the equilibrium in place.

The usual suspects are real enough: unsafe transport, poor working conditions, informality, scarce jobs. But there is a quieter constraint that deserves more attention, because it costs nothing to maintain and almost nothing to remove. It is a mistake, held collectively, about what other people think. A society can hold itself in an equilibrium that almost no one privately wants. Each person goes along with it because each believes the others prefer it, and no one acts in a way that would reveal the belief to be false. The question is not what Pakistanis think about women working. It is what Pakistanis think other Pakistanis think.

In a new study, my co-authors and I surveyed more than 8,000 men and women from low-income households in Faisalabad, Lahore and Sialkot, all living within commuting distance of apparel factories. We asked each person two things: do you yourself approve of women working outside the home, and how many out of 10 men, and 10 women, in your community do you believe approve?

The belief most strongly associated with a woman’s search and work history is her own perception of other women’s approval.

The first answer is striking on its own. Seventy-one per cent of the women we surveyed personally approve of women working outside the home. Yet only 18pc are working. The second answer explains much of the gap. Support for women’s work is widely underestimated, especially support among women, so communities are substantially more accepting than their members believe. On average, respondents understate men’s support modestly, but they understate women’s support by 14 percentage points, and the single largest error is men’s underestimation of other women’s support. The group that most approves of women’s work is the group everyone believes disapproves.

Which of these mistaken beliefs actually keeps women at home? The famous evidence from Saudi Arabia pointed at men: husbands privately supported their wives working but underestimated other men’s support, and correcting that one belief increased job search. Men matter in Pakistan too. In our data, the male household head’s beliefs about other men predict whether a woman currently works, consistent with his role in granting permission.

But the belief most strongly associated with a woman’s search and work history is her own perception of other women’s approval. A woman who underestimates how many other women support work is about a third less likely to have ever searched for a job. And this pessimism does not stay in her head. Women who hold it know less about how factory jobs work, what they pay, how to apply, and their own approval of women’s work is some 15 percentage points lower. What looks from outside like preference is often inherited pessimism: mothers’ beliefs about women predict daughters’ beliefs about women, passed down the female line like a family heirloom nobody asked for.

The evidence points most strongly to social learning about respectability rather than fear of punishment. A woman is primarily reading, from the imagined disapproval of other women, whether paid work is something a woman like her can respectably do. And because she reads the community as more disapproving than it is, she withholds the very behaviour that would reveal the error. So does her neighbour. The silence of each becomes the evidence for all.

This framing matters for policy, because preferences are difficult to change, but mistaken beliefs can be corrected. We have spent decades, rightly, on schools, transport and skills. Those investments moved. Participation did not. If a shared, false belief about social approval is part of what remains, then simply revealing that most women, and many men, already approve, is among the cheapest interventions available to the state, to employers, and to media.

Our findings also say whose beliefs should be corrected. Not only men’s. The women themselves, and the older women who form and pass on those beliefs. The case is strongest for women who are less connected; for them, information stands in for a network they do not have. And because these beliefs travel from mother to daughter, correcting one woman’s view may loosen the constraint on her daughter’s.

None of this is a substitute for jobs, safe transport or decent working conditions. But it does mean that part of what holds back half of Pakistan’s talent is not a wall but a mirage. Pakistan’s women are being held back, in part, by a picture of Pakistan that Pakistanis themselves do not recognise as their own.

The writer is an associate professor of economics at the Lahore School of Economics.

Published in Dawn, July 15th, 2026 ...