Invisible Women: Exposing Data Bias in a World Designed for Men

by Caroline Criado Perez

  The US is not alone in having a tax system that, by failing to account for gender, ends up discriminating against women. A recent paper expressed bafflement at how ‘many OECD countries’ were passing legislation in an attempt to reduce the gender pay gap while at the same time effectively increasing it through their family tax and transfer systems.24 Two such countries are the UK and Australia where, although married couples file separate income tax returns, most benefits and tax credits still breach the principle of independent taxation.

  The UK’s Marriage Allowance gives the main wage earner (usually the man) a tax break in couples where the lower earner is on £11,500 or less.25 This bolsters the gender pay gap on two fronts: supplementing male income, while also creating a perverse incentive for women to work fewer paid hours. Japan has a similarly male-biased married-couples tax break. Since 1961, the ‘head of household’ (normally a man) has been able to ‘claim a tax deduction of ¥380,000 ($3,700) as long as his spouse’s income does not exceed ¥1.03m (around $10,000)’. A 2011 survey by Japan’s labour ministry found that ‘more than a third of married women who worked part time and deliberately curtailed their hours did so to keep the tax deduction’.26

  In a slightly different example of a hidden gendered bias, Argentina’s tax system provides a rebate almost four times higher for employees than for the self-employed. Gender comes into it because men are more likely to be employed in the formal economy, while women are more likely to be self-employed in the informal economy.27 So the tax system is essentially covertly giving a higher rebate to men than to women.

  There’s a fairly simple reason why so many tax systems discriminate against women, and that is that we don’t systematically collect data on how tax systems affect them. In other words, it’s because of the gender data gap. The impact of taxation on women is ‘an underdeveloped area of research’ according to a 2017 report from the European Parliament, which called for more sex-disaggregated data on the issue.28 Even countries such as Spain, Finland and Ireland that have taken steps to analyse their budgeting from the perspective of gender, usually focus only on spending, not tax. In the EU, Austria ‘is one of the few countries where the government has defined specific goals for the tax system, such as promoting a more equal division of paid and unpaid work between women and men, enhancing the labour participation of women and reducing the gender pay gap’. Meanwhile, a 2016 survey of EU member states found that only Finland and Sweden have strictly individualised income tax systems.29

  The tax system’s woman problem extends beyond the zombie assumption that household resources are allocated equally between the sexes: it encompasses the theory of taxation itself – at least in its current form. Since the 1980s, governments around the world have been less interested in taxes as a means to redistribute resources, seeing tax more as a potential retardant to growth that must be contained. The result has been lower taxes on capital, corporations and high-income earners, and an increase in loopholes and incentives so that multinational corporations and the super-rich can avoid and evade tax. The idea is not to ‘distort otherwise efficient market processes’.30

  When gender has come into this framework at all, it has been solely in the context of how tax might harm growth by disincentivising women to enter paid employment. What isn’t considered is how a tax system focused so narrowly on enabling ‘growth’ benefits men at the expense of women. Cuts in the top rates of income tax disproportionately benefit men because of the gender pay gap. For the same reason, the majority of women in the world are not in a position to make use of the various tax loopholes an expensive accountant can afford you. Decreases in (or non-enforcement of) wealth and asset taxes also disproportionately benefit men, because men are far more likely to control such resources.31

  But it’s not just about benefiting men over women. These male-biased benefits actually come at women’s expense, because as we’ve seen, women have to fill the resulting service gaps with their unpaid care work. In 2017, the Women’s Budget Group pointed out that at the same time that austerity measures were having a particularly severe impact on women in the UK, ‘tax giveaways disproportionately benefitting men will cost the Treasury £44bn per annum by 2020’.32 These include a £9 billion cut in fuel and alcohol duties, a £13 billion cut in corporation tax, and a loss of £22 billion from raising income tax and National Insurance thresholds. Together, these tax giveaways accounted for more than the total annual cuts in social security spending – which makes it clear that this isn’t a matter of resources, so much as (gendered) spending priorities.

  The problem of low tax revenues in low-income countries is exacerbated by cross-border tax-avoidance techniques: multinational companies often ‘negotiate tax holidays or incentives as a condition for bringing their business to developing countries’, costing developing countries an estimated $138 billion in revenue annually. Well, the argument goes, if massive corporations paying zero taxes while they exploit cheap labour is the only way to get them there . . . Only it isn’t. The OECD has found that ‘such incentives are rarely a primary reason for investment in developing countries’.33 Women’s cheap labour, on the other hand, is certainly quite the draw. Nevertheless, such tax systems are sometimes ‘imposed as conditions on developing countries by international financial institutions’.34

  In a parallel to UK tax giveaways that outpace its spending cuts, the IMF estimates that developing countries lose $212 billion per year from tax-avoidance schemes, which far outstrips the amount they receive in aid.35 Over a third of the world’s total unrecorded offshore financial wealth is thought to be secretly held in Switzerland, which recently faced questions from the UN ‘over the toll that its tax and financial secrecy policies take on women’s rights across the globe’.36 A 2016 analysis by the Center for Economic and Social Rights (CESR) found that the amount of money lost to tax dodging by multinational copper firms such as the Swiss-headquartered Glencore in Zambia, could finance 60% of the country’s health budget. CESR also estimated that the Indian government lost out on up to ‘$1.2 billion in direct tax revenue from the funds held in just one bank branch in Switzerland – comparable to as much as 44% of [India’s] expenditure on women’s rights, and 6% of total social spending in the country in 2016’.37

  Governments need money, so they have to make up these losses somehow. Many of them turn to consumption taxes because they are easy to collect and difficult to evade. Low-income countries raise ‘about two-thirds of their tax revenue through indirect taxes such as VAT, and just over a quarter through income taxes’.38 A recent International Labour Organization analysis found that 138 governments (ninety-three developing and forty-five developed countries) are planning to either increase and/or extend consumption taxes, primarily through VAT.39

  This increase disproportionately affects women too. Not just because they are over-represented among the poor (the poorer you are the higher a proportion of your income goes on consumption), but also because they tend to bear the responsibility of buying food and household goods. And because women’s paid labour supply is more elastic (in no small part because of the gender pay gap), increasing VAT can have the effect of pushing women to spend more time in unpaid work in order to produce in the household what they might otherwise buy on the market.

  This problem is exacerbated by an often gender-insensitive allocation of what products do and don’t have VAT added, driven by an overall lack of research based on sex-disaggregated data on the impact specific consumption tax rates and exemptions have.40 VAT is not generally added to products that are seen as ‘essential’, so in the UK, food is exempt because it’s considered essential, while iPhones are not because they are not. But one man’s frivolity is another woman’s essential, and around the world women have been campaigning to get male-dominated legislators to recognise that sanitary products are not luxury items. In some countries they’ve even succeeded.

  It’s clear that tax systems around the world, presented as the objective trickle-down of market-driven fo
rces have intensely gendered impacts. They have been created based on non-sex-disaggregated data, and male-default thinking. Together with our woman-blind approach to GDP and public spending, global tax systems are not simply failing to alleviate gendered poverty: they are driving it. And if the world cares about ending inequality, we need to adopt an evidence-based economic analysis as a matter of urgency.

  CHAPTER 14

  Women’s Rights are Human Rights

  What the past two chapters have shown is that there are substantial gender data gaps in government thinking, and the result is that governments produce male-biased policy that is harming women. These data gaps are in part a result of failing to collect data, but they are also in part a result of the male dominance of governments around the world. And while we may not think of male-dominated government as a gender data gap problem, the evidence makes it clear that female perspective matters.

  Several US studies from the 1980s to the 2000s have found that women are more likely to make women’s issues a priority and more likely to sponsor women’s issues bills.1 In the UK, a recent analysis of the impact female MPs have had in Westminster since 1945 found that women are more likely to speak about women’s issues, as well as family policy, education and care.2 An analysis3 of the impact of female representation across nineteen OECD countries4 between 1960 and 2005 also found that female politicians are more likely to address issues that affect women.

  The OECD study also found that women’s words translated into action. As female political representation increased in Greece, Portugal and Switzerland, these countries experienced an increase in educational investment. Conversely, as the proportion of female legislators in Ireland, Italy and Norway decreased in the late 1990s, those countries experienced ‘a comparable drop in educational expenditures as a percentage of GDP’. As little as a single percentage point rise in female legislators was found to increase the ratio of educational expenditure. Similarly, a 2004 Indian study of local councils in West Bengal and Rajasthan found that reserving one-third of the seats for women increased investment in infrastructure related to women’s needs.5 A 2007 paper looking at female representation in India between 1967 and 2001 also found that a 10% increase in female political representation resulted in a 6% increase in ‘the probability that an individual attains primary education in an urban area’.6

  In short, decades of evidence demonstrate that the presence of women in politics makes a tangible difference to the laws that get passed. And in that case, maybe, just maybe, when Bernie Sanders said, ‘It is not good enough for someone to say, “I’m a woman! Vote for me!”’, he was wrong. The problem isn’t that anyone thinks that’s good enough. The problem is that no one does. On the other hand plenty of people seem to think that a candidate being a woman is a good enough reason not to vote for her. Shortly before the 2016 US presidential election, the Atlantic published the results of a focus group of undecided voters.7 The main takeaway was that Hillary Clinton was just too ambitious.

  This is not a groundbreaking opinion. From Anne Applebaum (‘Hillary Clinton’s extraordinary, irrational, overwhelming ambition’8), to Hollywood mogul, democratic donor and ‘one-time Clinton ally’9 David Geffen (‘God knows, is there anybody more ambitious than Hillary Clinton?’10), via Colin Powell (‘unbridled ambition’11), Bernie Sanders’ campaign manager (‘don’t destroy the Democratic Party to satisfy the secretary’s ambitions’12), and, of course, good old Julian Assange (‘eaten alive by her ambitions’13), the one thing we all seem to be able to agree on (rare in this polarised age) is that Hillary Clinton’s ambition is unseemly. Indeed, so widespread is this trope it earned itself a piece in the Onion headlined, ‘Hillary Clinton is too Ambitious to be the First Female President’.14

  Being the first woman to occupy the most powerful role in the world does take an extraordinary level of ambition. But you could also argue that it’s fairly ambitious for a failed businessman and TV celebrity who has no prior political experience to run for the top political job in the world – and yet ambition is not a dirty word when it comes to Trump.

  Associate professor of psychology at UC Berkeley Rodolfo Mendoza-Denton has a cognitive explanation for why we may view Clinton’s ambition as ‘pathological’.15 She ‘was forging into a territory that is overwhelmingly associated in people’s minds with men’. As a result, he explains, voters experienced her candidacy as a norm violation. And norm violations are, Mendoza-Denton writes, ‘quite simply, aversive, and are often associated with strong negative emotion’.

  There’s a very simple reason that a powerful woman is experienced as a norm violation: it’s because of the gender data gap. I personally grew up heavily buying into the myth that women are just . . . a bit rubbish. Yes, this was partly because that’s how women are represented in the media (consumerist, trivial, irrational) but it’s also because women are so under-represented. I was one of those girls being taught, via a curriculum, a news media and a popular culture that were almost entirely devoid of women, that brilliance didn’t belong to me. I wasn’t being shown women I could look up to (either past or present). I wasn’t being taught about female politicians, female activists, female writers, artists, lawyers, CEOs. All the people I was taught to admire were men, and so in my head power, influence, and ambition equated with maleness. And, if I’m being really honest, I think I experienced this norm violation as well. I was all too ready to accept the idea that female bosses were just too ambitious – which as we all know is code for bitch.

  The unpalatable truth is that it is still considered unladylike for a woman to want to be president. A 2010 study found that both male and female politicians are seen as power-seeking, but that this is only a problem for female politicians.16 In a similar vein, Mendoza-Denton conducted a study which found that context determines how ‘assertive’ men and women are judged to be.17 In a stereotypically ‘male’ context (car mechanic, Wall street, president of the United States) a woman is judged to behave more assertively than a man saying exactly the same as her. And while it was OK if a bit odd for men to be assertive in a ‘female’ context (choosing curtains, planning a child’s birthday party), it was definitely not OK for a woman to be assertive in any context. Assertive women are bossy.

  The social downer on women being seen to seek professional power is partly because social power (being seen as warm and caring) is women’s ‘consolation prize for renouncing competition with men,’ write psychology professors Susan Fiske and Mina Cikara.18 Social power for women is therefore intrinsically incompatible with professional power: if a woman wants to be seen as competent she has to give up being seen as warm.

  But so what. So you’re disliked. So you’re seen as cold. Suck it up. If you don’t like the heat, get back to the kitchen, right?

  Wrong. That would be to assume that men face the same heat for being seen as cold. They don’t. The 2010 study didn’t just find that female politicians were see as less caring. It found that this perception inspired moral outrage in both male and female study participants, who viewed such women with contempt, anger and/or disgust. This was not the case for their male counterparts. Molly Crockett, associate professor of experimental psychology at Oxford University, has an explanation for this disparity: being seen as uncaring is a norm violation for women in a way that it just isn’t for men. ‘There is an expectation’, she tells me, ‘that on average women are going to be more pro-social than men.’ Any deviation by a woman from what is seen (no matter how illogically) as a ‘moral’ stance therefore shocks us more.

  Given the clear significance of gender when it comes to these issues, you would hope that this might be an area of research that bucks the gender-data-gap trend. It does not. Imagine my excitement when I came across a paper published in January 2017 entitled ‘Faced with exclusion: Perceived facial warmth and competence influence moral judgments of social exclusion’.19 Given the findings of Fiske and Cikara about women’s warmth/competence trade-off this should have been an extremely useful paper.
As the authors explain, ‘people’s moral judgment about social exclusion can be influenced by facial appearance, which has many implications in intergroup research’. That is, people’s decisions about whether or not it’s fair that someone is being ostracised or bullied can be influenced by what the victim looks like.

  Indeed. Unfortunately, the study authors ‘used male faces only for reasons of test efficiency’, making the study absolutely worthless when it comes to the group most affected by this issue, i.e. women. Fiske and Cikara explain that gender, ‘is a salient, and perhaps the most salient, social category’, with gender stereotyping often being immediate and unconscious: ‘the mere sight of a woman can immediately elicit a specific set of associated traits and attributions, depending on the context’. Still, at least the test was efficient.

  ‘It’s actually kind of shocking how little attention there’s been to gender in the morality literature,’ says Crockett. But on the other hand, maybe it’s not: the study of morality, Crockett tells me, is ‘really aiming at trying to uncover human universals’. At the point she mentions ‘universals’, of course, male-default-thinking alarm bells start ringing in my head. Many academics in the field of morality subscribe to ‘very egalitarian, utilitarian, impartial views of what is right’, Crockett continues, and they perhaps impose those norms ‘onto the research that we do’. The alarm bells ring off the hook.