Rejection matters

“Rejection rates in employment are huge,” says Isabel Fernandez-Mateo. “In many recruitment processes, around 95% of people get rejected. Over the course of their career, people are often passed by for promotion during their journey to the top. If, as some research has shown, men and women react differently to those kinds of rejections, this could lead to gender disparities in advancement. Rejection matters.”

For Isabel, Adecco Professor of Strategy and Entrepreneurship at London Business School, rejection means much more than its negative connotations. Her new research paper, written with Brian Rubineau and Venkat Kuppuswamy, reveals its impact on diversity. Titled Reject and Resubmit: A Formal Analysis of Gender Differences in Reapplication and Their Contribution to Women’s Presence in Talent Pipelines, the paper challenges popular assumptions and reaches surprising conclusions. 

“Women are underrepresented in top management, entrepreneurship, science and in many other contexts,” she begins. “One common explanation for this is that they exit talent pipelines at higher rates than men. We also know that when men and women seek a particular opportunity, and get rejected, men are more likely to try again for that same opportunity than women. Our paper is about checking whether this phenomenon will have important long-term consequences for gender diversity in talent pipelines.” 

Two sides of the equation

A native of Spain, Isabel has been studying gender diversity for many years, since obtaining her PhD from the MIT Sloan School of Management. An expert on how relationships influence career outcomes, she teaches two elective courses at LBS – “Building your Career Strategy” and "People Analytics” – as well as teaching in the PhD programme.

In a paper she wrote on rejection a few years ago, she discovered that women were not necessarily risk-averse or more sensitive to rejection than men, but that many chose not to reapply for a job because they felt the hiring process was unfair. Isabel looked into this from both sides of the equation – from the applicants’ behaviour (the “supply side”) and from the job-screeners’ behaviour (“the demand side”).

On the ‘demand side’, some research has shown that there can be discrimination in hiring, biassed promotion processes, and unwelcoming cultures. On the ‘supply side’, women may choose not to apply for certain jobs because they have different preferences than men. However, the supply and demand side are not independent of each other – they interact and feed into each other, affecting the outcome. For example, women may decide not to apply to jobs because they anticipate they’re not going to get hired, whether that’s true or not. Or they could think that the culture will make them feel like they don’t belong. Isabel’s research suggests that women’s perceptions of the fairness of the hiring process (rather than their intrinsic preferences) are a key determinant of whether they chose to apply for a job. 

Women’s behaviour

For her latest paper, Isabel looked into the diversity implications of these findings in three male-dominated settings: executive search, patenting and crowdfunding. “These are fields in which there are very few women, which is important, because diversity matters, right?” In challenging popular, targeted interventions to address women’s under-representation in talent pools, a surprising finding emerged.

“An instinctive reaction of organisations when trying to decrease the gender gap in these situations is to convince women to reapply – i.e., change women’s behaviour. They say, ‘Oh, 80% of our applicants are men, let's make it known that we’re a women-friendly organisation and get more women applying.’ But we discovered that the most obvious, straightforward solutions aren’t always the most effective. Perversely, they can actually increase the segregating effects, and backfire.

“The more people you have in the application pool, the more you will reject. And the higher the rejection rate, the fewer women there will be in the application process in the long run.” 

“Women can be convinced to reapply, but in contexts where the rejection rate is very, very high – such as hiring for executive positions – acting on that lever will be very hard. What we show is that women’s behaviour does matter, but sometimes not as much as the company’s behaviour.”

The crowdfunding angle

Isabel’s research also challenges the assumption that, because women are not reapplying for opportunities, there will be fewer and fewer women in the talent pipeline in the long run. “We asked, is that true? Well, we found in certain settings it is true; in some it isn’t; and in others it just doesn’t matter.”

Crowdfunding, she found, is one setting where they already do much cooler things to keep rejection rates low. “They give training to people, so they can give better pitches. And crowdfunding is one of the few entrepreneurial settings where women are more likely than men to obtain funding. You might spend a lot of money trying to convince more women to re-apply for crowdfunding, but this practice won’t necessarily impact diversity – you might as well spend the money on other things to help diversity.”

Diversity initiatives

“The big problem in diversity initiatives,” stresses Isabel, “is that people don’t stop to look at the context and the levers they already have – they use the thing everybody is doing because they assume that will solve the problem. But the problem might be a different problem to what they think it is. 

In rethinking how to promote diversity, Isabel believes there are other policies, interventions and practices which will be effective. What are they?

“Lowering the rejection rate is an important one,” she says. “And try to lower it for everybody – not just for men and women – but with a gender-neutral policy that (surprisingly) will end up increasing diversity. If you lower it, for example, from 90% to 60% for everybody, you’ll be rejecting fewer applicants, which in turn increases diversity.

“The question is, of course, how do you go about lowering the rejection rate? You can’t just say, ‘Let’s accept everybody and be more lenient,’ because that makes no sense. The selection process exists for a reason.”

One method of lowering the rejection rate, she suggests, is for organisations to provide realistic job previews which give an accurate picture of the position; thus, only the more suitable applicants will apply. Likewise, lowering applicants’ search costs can also be instrumental, as they impact perceptions of job fit and will only encourage applications from candidates who are less likely to be rejected.

Organisations can even eliminate the need to reapply by having candidates consent at the time of application to be considered for other relevant opportunities that might come along in the near future. Job tryouts and internships should also be considered, as they give a better understanding of how the company operates and have similar effects on lowering rejection rates for certain positions. 

Meanwhile, tailoring rejection messages – such as providing feedback after an unsuccessful interview or application – can still be an effective lever in reducing the gender gap. 

“A recent study found that women who were rejected with messages citing ‘fit’ with the job (i.e., their suitability to the position) were significantly more likely to reapply for jobs than those who were rejected with messages citing quality or were rejected without a reason,” Isabel says. 

Sex bias analysis

Another surprising find of the research was that even if the rejection rates are unbiased, this can impact the segregating effects of job reapplication – a possibility that has not been considered in previous studies.

“People often think that differences in representation are to do with gender bias, but it’s not a given that women are more likely to be rejected; men are equally likely to be rejected in certain scenarios. It often comes down to the numbers. Take a case where 50% of women and 50% of men are rejected vs a case where 90% of women and 90% of men are rejected – in the latter case the diversity implications will be much stronger over time because more people are being rejected. So, even if there’s no hiring bias, the rejection rate is going to matter.

A neat model

Because researchers cannot directly evaluate interventions that boost future diversity in the real world, Isabel used a novel mathematical modelling approach to simulate various scenarios that might arise. 

“The real world is a very messy reality,” she says. “You can’t look into the future to see what’s going to happen, because you often don’t have the data to test things. That’s why we have a model. You can use mathematics to give an indication that something in the real world, that we might not have considered before, might actually be going on. It’s a very neat method”.

“So many things happen in these scenarios, and they all interact with each other. The women are not applying; the organisation is biassed; the rejection rate is high; there aren’t enough women with the right qualifications. All these variables go into determining that few women are reapplying. The problem is, how do you prove this?” 

By putting all these variables into a mathematical equation, Isabel and her team found how changing one thing, while keeping everything else the same, would affect the result. 

“You can’t do this in the real world,” she says, “but in a formal model we can. Let’s say that women are rejected at the same rate, and if we change this, this is what happens. Or if we change the likelihood of women reapplying, this is what happens. It could even show that if we introduced a new policy in the real world, there could be unpredictable and unintended consequences.”

Looking ahead

Isabel believes the time is ripe to open up new avenues for research on gender disparity in the reapplication process. But she also wants to make one thing very clear.

“I don’t want to say, ‘Women, you have to do this.’ There’s nothing that women have to do here. There’s a lot of talk about women in business and all the things they need to do to get ahead. This paper is not about that.” What she does hope is that organisations will pay more attention to how they structure their selection process.

“It’s important to look at how they reject people, how many they reject, how many people are in application pools… All these things, even if they don’t immediately seem to be a diversity policy, will affect how many women will be in business in the long run.” 

She also hopes that companies will consider the outcomes of any diversity practices they put in place more carefully; and not just introduce them because that’s what everyone is doing right now. “They might think, ‘Let’s increase the application pool’, ‘let’s do diversity training’, ‘let’s do implicit bias training’, ‘let’s recruit a diversity officer’... But companies need to work out if that’s something that will matter in their context and if it will make a difference in the long run. The consequences could be counterproductive.” 

“It’s how we sort through this process, the practices we put in place, that will affect how many people apply for positions – whether they stay, whether they make it to the top, how many women make it to the top, and so on. That is what’s important.”

Isabel Fernandez-Mateo is Adecco Professor of Strategy and Entrepreneurship, Strategy and Entrepreneurship Faculty at London Business School