My recent research (joint with Warn Lekfuangfu) considers the extent to which societal shifts have been responsible for an increased tendency for women to sort into traditional male roles over time, versus individual level childhood factors (for example, socioeconomic status, parental time inputs, peer effects). In other words, we are interested in the extent that childhood factors, which vary within cohorts, influence the type of job selected as compared to societal changes, which occur across cohorts.

We investigate this question using three large cohort studies in the UK that are made up of children born in 1958, 1970 and 2000. We find very strong evidence that societal movements have done the heavy lifting with respect to changing occupation choice across genders. We also find that children aspire to jobs along gendered lines for all three cohorts, despite substantial progress being made over time. In other words, gender representation is still not 50:50 for occupation choice in the most recent cohort.

Our analyses also revealed a couple of interesting stylised facts. First, for all three cohorts we find persistent gender gaps in the tendency to sort into occupations with the highest shares of males (>=80 per cent) that have not changed over time. This is true for both the average and highest ability child. For the highest ability children, these jobs are often the golden pathway to C Suite positions and positions of power, and encapsulate science, technology and engineering posts as well as front office trading roles and politics.

It may be tempting to conclude that the gender gap flatness in the tendency to sort into occupations with the highest share of males, particularly for children with the highest academic ability, reflects innate preferences. However, we note that over time both genders have significantly changed their tendency to sort into occupations that differ in terms of other activities and context. Some of this will be determined by labour markets (i.e., it is unsurprising that both genders sort towards jobs that are high in people, for example, given the growth in services and jobs that require interpersonal skills), but we also view these changes as highly suggestive that preferences are socialised, rather than innate. This is why societal movements are so important for noticeable changes in occupation choice.

Second, while all eyes are normally on the tendency for females to change their preferences, noteworthy is that the males in the most recent cohort (currently about 19 years old, so they are our next generation of professional workers) are aspiring to occupations with significantly higher levels of competitiveness and larger incomes as compared to previous cohorts and their current female peers. Therefore, the gender pay gap may prevail, unless the rewards given to different occupations change (i.e., occupations with high shares of females, like nursing, garner higher pay), or indeed preferences for these young males are not realised. This finding underlines the risks of over-focusing on getting females into traditionally male careers, but not paying attention to the changing trends – or lack of –of preferences for males.

Our study raises questions on what can really be achieved by individuals at a local level and by parents to move the needle on gendered sorting in the absence of a more general societal movement or a tipping phenomenon. For example, if a mother encourages their daughter to be an astrophysicist, but the society she is growing up in sends different messages, the efforts may be lost on the average girl. It is possible that these messages may be dominated by, for example, STEM toys being mainly targeted to boys, the media covering females and males at the height of their careers differently, a child’s schooling experiences varying by their gender and the images society has for its leaders still being male. Overall, we view our work as underlying the importance of the role of societal shifts, over and above childhood variables, in determining the sorting patterns we have seen over the last number of decades in the UK, and also those that remain today. There is some truth in the saying that it takes a village!

Currently, there is a lot of pressure on firms globally to address gendered sorting, and hire more females into traditionally male occupations. This has led some governments to mandate gender quotas for company boards, for example Norway, Netherlands, Israel and France. Elsewhere, global finance firms are committing to gender targets without being legally obliged to do so: Barclays, Royal Bank of Scotland, HSBC and Lloyds have all signed ‘the Women in Finance Charter’ and promised specific levels of progress by hard deadlines. In Silicon Valley, big tech firms, like Twitter and Pinterest, have also publicly declared similar non-mandated targets (and subsequently failed to meet them — for example Facebook, Pinterest and Twitter (2017).

The field of law has seen similar proclamations, with major international law firms like Herbert Smith Freehills and Pinsent Masons setting ambitious targets. This pressure implies that it is the characteristics of work culture that cause women to avoid certain occupations, implying that firms can solve the problem of 50:50 male-female representation by changing their policies contemporaneously.

However, our work points also to gender norms in occupational choice. At the firm level, policy responses then involve more than just firm-level characteristics. Assuming, gender norms are socialised, downstream education policies which de-sex occupation choice, and firm level policies which put females in visible roles may change the gender norms that affect occupational choice. Next to this, our work underlines the importance of addressing the gender norms carried by males in society. Without this attention, the patterns observed in our study are unlikely to get fully eroded.



Grace Lordan is an associate professor in behavioural science at LSE. She is an economist by background, and her research is focused on understanding why some individuals succeed over others because of factors beyond their control. In this regard, she has expertise on the effects of unconscious bias, discrimination and technology changes. Grace is also interested in using the techniques of behavioural science to design interventions for firms to promote good conduct, diversity and inclusion and curb biases that creep into high stakes decision-making. Grace has advised and given talks to large investment banks and international conferences on these topics. Grace has also led projects to advise commissioners in the UK and policy makers in the EC. At the LSE Grace trains executives in these approaches through her teaching on corporate behaviour and decision-making.